Resources

ABLags generates a VECT[SERIES] with the expanded panel data instruments for doing Arellano-Bond instrumental variables estimation. Arellano, M. and S. Bond (1991). "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations.” Review of Economic Studies, Vol. 58, 277-297. See also arellano.prg.

ACF performs an autocorrelation analysis on a series. This is a simpler routine than the related BJIDENT and REGCORRS procedures.

This computes a series of partial autocorrelations from a series of autocorrelations.

Takes as input a set of AR and MA coefficients, and generates the corresponding theoretical autocorrelations. Updated November 10, 1997, it can now produce autocovariances as well, and allows skipping beginning and ending items in the input vector. (Note: This procedure file was renamed from ACF.SRC to ACFMORIN.SRC in July, 2005 to avoid confusion with a newer ACF.SRC written by Estima and included with RATS starting with Version 6.10)

Demonstrates adaptive kernel estimation: a two-step procedure for adjusting the estimates of a linear model based upon the an estimated density for the residuals. Uses the DENSITY instruction for computing the density and MCOV instruction for computing the adjustment matrices.

This selects the optimal lag length for an ADF unit root test (it does not do the UR test itself--see the Unit Root section for testing proceudres). This uses features of RATS 6.0 to report the results in a convenient format. It is largely based on a procedure found in Norman Morin's URADF.SRC file. Note: The filename was renamed from ADFAUTO.SRC (to match the name of the procedure itself) in August, 2005.

Replication file for Aruoba, Diebold and Scotti(2009), "Real-Time Measurement of Business Conditions," Journal of Business and Economic Statistics, vol 27, no 4, 417-427.
Instead of brute force with a huge state vector, this uses aggregator states to handle the lags of the indicator for the two flow variables. The authors have an update on the ADS website that describes this.

This implements the Anderson-Darling test for normality.

Estimates the fractional difference power for a series using the bias-reduced technique from Andrews and Guggenberger(2003), "A Biased-Reduced Log-periodogram Regression Estimator for the Long-Memory Parameter", Econometrica, vol 71, no. 2, 675-712.

Calculates and displays Akaike Information Criteria and Schwarz Bayesian Criteria for distributed lag models.

Andrews-Ploberger, Andrews-Quandt tests for structural break in a linear regression using Hansen's p-values. Andrews, Donald W. K. and Werner Ploberger, "Optimal Tests When a Nuisance Parameter is Present Only Under the Alternative", Econometrica, 1994.

Andrews-Ploberger, Andrews-Quandt tests for structural break in a general maximum likelihood.
Andrews, Donald W. K. and Werner Ploberger, "Optimal Tests When a Nuisance Parameter is Present Only Under the Alternative", Econometrica, 1994.

Estimates a linear regression model with AR1 errors using a variety of techniques (Cochrane-Orcutt, Hildreth-Lu, maximum likelihood).

Computes an information criterion for various lags of AR processes using the Burg or Yule estimates of the partial autocorrelations. Identifies the best choice for the given criterion.

Tests for ARCH effects in a series. Can do a single test for a set number of lags, or a sequence with different numbers of lags.

Estimates a dynamic panel data model using Arellano-Bond GMM estimation. Arellano, M. and S. Bond (1991). "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations.” Review of Economic Studies, Vol. 58, pp. 277-297. See also ablags.src

Replication file for Arellano and Bond(1991), "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations", Review of Economic Studies, vol 58, no. 2, 277-97.

Procedure for generating true simulations from ARFIMA(0,d,0) using the Davies and Harte Method. Written by Rob Schoen. You can download the Zipped file ARFSIM.ZIP which includes the ARFIMA simulation procedure, a required Gamma function procedure, and a short example progra, or download the individual text files: ARFSIM.SRC, XGAMMA.SRC, and ARFSIM.PRG (example program).

Demonstrates simulation of a univariate (conditionally Normal) AR(1)-GARCH(1,1) model

Estimates, tests and forecasts ARIMA models. Demonstrates the BOXJENK and UFORECAST instructions, BJTRANS, BJIDENT, BJAUTOFIT and REGCORRS procedures.

Sets up the transition equation matrices for a state-space representation of an ARMA mode. The procedure takes an input ARMA equation you supply and sets up appropriate "A" and "SW" matrices.
Jones(1980), "Maximum Likelihood Fitting of ARMA Models", Technometrics, 389-395.

Example of analysis of an ARMA model using Gibbs sampling (Independence Chain Monte Carlo).

This produces a graph of the spectral density for an input ARMA model, where the model is in the form of an equation. (Renamed and revised from the older ARMASPEC.SRC file).

Uses @BJDIFF and @GMAUTOFIT to do an automated choice for the differencing and seasonal ARIMA specification for a series, then uses BOXJENK to estimate the chosen model, allowing for automated outlier detection.

Replication file for Bai, Lumsdaine and Stock(1998), "Testing For and Dating Common Breaks in Multivariate Time Series", Review of Economic Studies, vol 65, no 3, 395-432
Calculates break statistics on a multivariate models with a common break across all equations.

Replication file for empirical examples from Bai & Perron(2003), "Computation and analysis of multiple structural change models," Journal of Applied Econometrics, vol. 18, no. 1, pages 1-22.

Replication file for Baillie and Bollerslev, "The Message in Daily Exchange Rates: A Conditional Variance Tale", JBES 1989, vol 7, pp 297-305 which estimates univariate GARCH models with day-of-the-week effects.

Replication of Baillie, Bollerslev and Mikkelson(1996), "Fractionally Integrated Generalized Autoregressive Conditional Heteroskedasticity", Journal of Econometrics, vol 74, pp 3-30.
Demonstrates the DIFFERENCE instruction with the FRACTION option.

Estimates the required number of factors in a linear factor model using the formulas in Bai and Ng(2002), "Determining the Number of Factors in Approximate Factor Models", Econometrica, vol 70, 191-222.

This does multiple structural change analysis as described in Bai and Perron (2003), "Computation and Analysis of Multiple Structural Change Models", Journal of Applied Econometrics, vol. 18, 1-22.

Replication of Balcilar, Gupta, Miller(2015), "Regime switching model of US crude oil and stock market prices: 1859 to 2013", Energy Economics, vol 49, 317-327.

Analysis of a Markov Switching Vector Error Correction Model (MSVECM).

Replication file for Balke(2000), "Credit and Economic Activity: Credit Regimes and Nonlinear Propagation of Shocks," Review of Economics and Statistics, vol 82, 344-349.
This estimates a threshold VAR and computes impulse responses using bootstrapping of residuals.

Replication of Balke and Fomby(1997), "Threshold Cointegration," International Economic Review, vol 38, no 3, 627-45.

Example of simple forecast procedures.

Replication file for Bauwens and Laurent(2005), "A New Class of Multivariate Skew Densities, With Application to Generalized Autoregressive Conditional Heteroscedasticity Models," JBES, vol 23, 346-354.

Tests series for a unit root using the Bayesian procedure outlined in C. Sims, "Bayesian Skepticism on Unit Root Econometrics", J. of Economic Dynamics and Control, 1988 no. 2/3.

Replication for Bernanke, Boivin & Eliasz (2005), "Measuring the Effects of Monetary Policy: A Factor-augmented Vector Autoregressive (FAVAR) Approach," The Quarterly Journal of Economics, vol. 120(1), pages 387-422.

Performs the battery of independence tests used extensively in Brockwell and Davis (2002), "Introduction to Time Series and Forecasting" 2nd ed, Springer.

Computes the Brock, Dechert, & Scheinkman Test (BDS) test for i.i.d., "A test of independence based on the correlation dimension" (working paper title). Included in Brock, Hsieh and LeBaron, "Nonlinear Dynamics,Chaos,and Instability", MIT Press, 1993, Chap. 2.

Replication files for Bernanke & Mihov(1998), "Measuring Monetary Policy", QJE, vol 113, no 3, 869-902 (monthly data calculations).
This includes maximum likelihood estimation of structural VAR's, Markov switching estimate of an SVAR and Monte Carlo integration of a just identified SVAR.

Returns a 2-vector with the two parameters for a Beta distribution with the given mean and standard deviation

Estimates beta coefficients for a large number of stocks. Demonstrates use of looping instructions.

Computes the "C" and "H" portmanteau statistics for autocorrelation and third order dependence, from Hinich, M. (1996). "Testing for dependence in the input to a linear time series model." Journal of Nonparametric Statistics, 6, 205-221.

Given unusual events that do not fit adequately into a specified model and that manifiest themselves as high leverage points or as outliers, BIF implements the Bounded Influence Function Regression, a robust estimation proposed by Krasker, Ku, and Welsch (1983). This method downweights the influence of potentially influential observations avoiding relatively large contribution to the values of the estimates.

Example of a bivariate HP filter.

BJAutofit chooses the minimum AICC or BIC model for the dependent variable. It uses maximum likelihood estimation to ensure that the estimates are done over a consistent time interval.

Generates a table of Schwarz criteria for various combinations of differencings and mean extractions, as a preliminary step in Box-Jenkins modeling.

Updated version of BJEST procedure included with RATS (for estimating ARIMA models). This version features more options and improved graphics.

Estimates and forecasts an ARIMA model

Computes and graphs autocorrelations and partial autocorrelations of a series and its differences to aid in choosing an ARIMA model

Replication file for Bjørnland, Hilde C. and Kai Leitemo (2009), "Identifying the Interdependence between US Monetary Policy and the Stock Market". Journal of Monetary Economics, vol 56, pp 275-282.
Demonstrates use of the SHORTANDLONG procedure on a model with five variables, including Monte Carlo integration of the impulse responses.

This computes Theil U statistics for a Box-Jenkins ARIMA model.

This does offset graphs of the levels, square root and log of an input series to help determine which preliminary transformation is appropriate.

Implements a band pass filter on a series using the methods of Baxter and King (1999) "Measuring business cycles: Approximate band-pass filters for economic time series", Review of Economics and Statistics, vol 81, no. 4, 575-593. Revised and updated by Estima, and renamed as BKFILTER (to avoid confusion with other band pass filtering procedures).

Replication for Blanchard and Quah(1989), "The Dynamic Effects of Aggregate Demand and Supply Disturbances", AER, vol 79, no. 4, pp 655-673. Demonstrates several topics in VAR's: historical decomposition, recovery of structural shocks, long-run restrictions.

BlockBoot is an alternative to the BOOT instruction which does block draws. You use the output "boot" SERIES[INTEGERS] to do the resampling that you require. The BOOT instruction in RATS version 6.30 and later includes options for doing the same thing (and more).

Recursive minimum unit root test. Banerjee, Lumsdaine and Stock(1992), "Recursive and sequential tests of the unit-root and trend-break hypothesis: theory and international evidence", JBES, vol. 10, 271-287.

Rolling minimum unit root test. Based largely on a 1992 paper by Banerjee, Lumsdaine, and Stock.

Sequential minimum unit root test. Based largely on a 1992 paper by Banerjee, Lumsdaine, and Stock.

Performs the Beveridge-Nelson decomposition on a single series. (See MVBNDECOMP.SRC for multivariate decomposition).
Beveridge and Nelson(1981), "A New Approach to Decomposition of Economic Time Series into Permanent and Transitory Components", J. of Monetary Economics, vol 7, 151-174. The algorithm used is from Newbold(1990), "Precise and Efficient Computation of the Beveridge-Nelson Decomposition of Economic Time Series," J. of Monetary Economics, vol 26, 453-7.

Replication file for Bollerslev and Mikkelson(1996), "Modeling and pricing long memory in stock market volatility", Journal of Econometrics, vol 73, pp 151-184.
This estimates FIGARCH and FIEGARCH models (fractionally integrated GARCH and fractionally integrated EGARCH)

Estimates a term structure from a set of bond yields. Demonstrates estimation with a function which can't be written in closed form.

Estimates a yield curve using a cubic spline approximation to the discount function, as described in McCulloch(1971), "Measuring the Term Structure of Interest Rates", Journal of Business, vol 44, pp 19-31.

Demonstrates bootstrapping of an ARMA Model.
Reference: David S. Stoffer and Kent D. Wall(1991), "Bootstrapping State-Space Models: Gaussian Maximum Likelihood Estimation and the Kalman Filter", Journal of the American Statistical Association, Vol. 86, No. 416, pp. 1024-1033

Example of parametric bootstrap of univariate autoregression

Example of bootstrapping an FMOLS estimate of cointegrating vectors. Based upon the technique described in Li and Maddala, "Bootstrapping cointegrating regressions", J. of Econometrics 80 (1997) pp 297-318

Example of bootstrapping linear model with estimated heteroscedasticity correction.

Used with MODES.SRC, this bootstraps series and tests against an alternative number of modes. Requires MBKERNEL.SRC (a stripped-down version of the KERNEL.SRC procedure). Also available are MBDEMO.PRG and GDPQ.RAT, a sample program and data file. Or download MODES.ZIP which contains all of the above.

Tests the mean of a sample using bootstrapping. Demonstrates the BOOT instruction.

Example of bootstrapping a spectral density. Algorithm from Franke and Hardle(1992) "On Bootstrapping Kernel Spectral Estimates", Annals of Statistics, vol. 20, no 1, 121-145.

Example of bootstrapping a simple VAR to get error bands on the impulse responses.

Does a parametric bootstrap (to get error bands for an IRF) of a VECM with known cointegrating vector.

Shows several methods for estimating a Box-Cox model by maximum likelihood. Demonstrates MAXIMIZE, NLLS with the JACOBIAN option and a grid search using FIND.

Tests the residuals from a least squares regression for presence of random effects using variants of the Breusch-Pagan LM test. It produces the standard BP test (which has chi-squared asymptotics), the Honda one-sided form (which has N(0,1) asymptotics) and the SLM (standardized Lagrange Multiplier) which is similar to the Honda test but uses small-sample corrections for the mean and standard deviation.

Alternate to MCVARDoDraws which does Monte Carlo draws for a 2-variable BQ factorization.
Blanchard and Quah(1989), "The Dynamic Effects of Aggregate Demand and Supply Disturbances", AER, vol 79, no. 4, pp 655-673.

Implements panel unit root tests from Breitung(2000), "The local power of some unit root tests for panel data", from Baltagi, Fomby, Hill (eds), "Nonstationary Panels, Panel Cointegration, and Dynamic Panels (Advances in Econometrics, Volume 15)", Emerald Group Publishing, pp.161-177

Implements the Bry-Boschan (NBER) Business Cycle Dating Algorithm. Can be applied to either monthly data (Bry-Boschan) or quarterly data(Pagan-Harding).
Bry and Boschan (1971). "Cyclical Analysis of Time Series: Selected Procedures and Computer Programs", NBER, New York.
Pagan and Harding (2002) "Dissecting the cycle: a methodological investigation", Journal of Monetary Economics, Volume 49, Issue 2, 365-381.

BSOPTION computes the value of an option using the Black-Scholes formula. See e.g. Campbell, Lo and MacKinlay, "The Econometrics of Financial Markets", Princeton University Press, 1997. This is the same procedure file that is included with RATS. Revised in March, 2007, to use options, rather than a list of parameters.

Replication of Burnside(1994), "Hansen-Jagannathan Bounds as Classical Tests of Asset-Pricing Models," Journal of Business & Economic Statistics, vol. 12, no 1, pages 57-79.

Builds a prior for an equation in a BVAR one element at a time. Each element is independent of the others and Normally distributed.

Uses repeated calls to @BVARBuildPrior to builds a "Minnesota" prior for a VAR model.

Estimates a dynamic model using the combination of DSGE and DLM. RATS Version 8, User's Guide, Example 10.4.

Replication file for Camacho(2011), "Markov-switching models and the unit root hypothesis in real U.S. GDP", Economics Letters, vol. 112, 161-164

Replication for Campbell and Ammer(1993), "What Moves the Stock and Bond Markets? A Variance Decomposition for Long-Term Asset Returns", J of Finance, vol 48, pp 3-37.

Computes the canonical correlations and related statistics for two sets of series, possibly conditioning on the third set.

Demonstrates process for selecting the prior in a Bayesian VAR. Example of the SPECIFY instruction in SYSTEM definition, including standard symmetric priors and a general prior.

Example of two-step estimates of various DCC models. This is the technique described in Cappiello, Engle & Sheppard(2006), "Asymmetric Dynamics in the Correlations of Global Equity and Bond Returns," Journal of Financial Econometrics, vol. 4, no 4, pages 537-572 applied to a different data set.

Replication file for Kai Carstenson(2006), "Stock Market Downswing and the Stability of European Monetary Union Money Demand", JBES, vol 24, number 4, pp 395-402. Includes procedure and data files. Demonstrates FM OLS techniques and a Hansen stability test.

Solves the Cass-Koopmans growth model.
RATS Version 8, Reference Manual, Example from DSGE

Demonstrates several forms of bivariate causality tests, including the standard lag exclusion Granger method, Sims and Geweke-Meese-Dent versions of the distributed lag method.

Replication files for Cecchetti and Rich(2001), "Structural Estimates of the U.S. Sacrifice Ratio," JBES, vol. 19, no 4, 416-427.
This examines three different small structural VAR's of the US economy and uses them to estimate the "sacrifice ratio" of GDP lost to disinflation. Using bootstrapping, it shows that the sacrifice ratio is poorly estimated when uncertainty in the VAR is taken into account.
svar2 is for the Cecchetti (2 variable) model
svar3 is for the Shapiro-Watson (3 variable) model
svar4 is for the Gali (4 variable) model

Program for the determination of the turning points of the time series, calculation of the duration and amplitude of the expansions and recessions. The expansion (recession) is defined as a timespan between a cyclical trough (peak) and peak (trough). The amplitude hence is the absolute distance between trough (peak) and peak (trough).

This implements a generalized version of the Baxter-King bandpass filter developed by Christiano and Fitzgerald(2003), "The Band Pass Filter", International Economic Review, vol. 44, no. 2, 435-465. (Renamed from BPASS.SRC in July, 2005, to avoid confusion with other band pass filter procedures).

Example of GMM with multiple conditions using NLSYSTEM.
Based upon the model from Chan, Karolyi, Longstaff and Sanders(1992), "An Empirical Comparison of Models of the Short-Term Interest Rate" Journal of Finance, vol 47, no 3, 1209-1227.

Replication file for Chan and Maheu(2002), "Conditional Jump Dynamics in Stock Market Returns", JBES, vol 20, no. 3, 377-389. This does constant and time-varying intensity (ARJI) models.

Replication file for Felix Chan & Michael McAleer(2003), "Estimating smooth transition autoregressive models with GARCH errors in the presence of extreme observations and outliers," Applied Financial Economics, vol. 13, no 8, 581-592.

Demonstrates several ways to compute a "Chow" test for structural break at a known point with two or more samples. Shows the use of multiple LINREG's for the split samples, and also the use of the SWEEP instruction with GROUP for more complicated sample splits.

This demonstrates a "Chow" tests for sample split at a known location using a single regression with variables dummied out over the interval.

Computes the multiple variance ratio statistic from Chow, K.V. and Denning, K. (1993) "A simple multiple variance ratio test", Journal of Econometrics, 58, 385-401

ChowLin distributes a series, changing the frequency to a higher one while maintaining the sum over each period, using the Chow-Lin(1971) or related procedure. The newer procedure disaggregate.src is a better choice.
Chow and Lin(1971), "Best Linear Unbiased Interpolation, Distribution and Extrapolation of Time Series by Related Series", Review of Economics and Statistics, vol 53, 372-375.
Fernandez(1981), "A Methodological Note on the Estimation of Time Series", Review of Economics and Statistics, vol 63, 471-478.
Litterman(1983), "A Random Walk, Markov Model for the Distribution of Time Series", JBES, vol 1, 169-173.

Example of Chow tests in linear model

Replication files for Chan, Karolyi, Longstaff and Sanders(1992), "A Empirical Comparison of Models of the Short-Term Interest Rate" Journal of Finance, vol 47, no 3, 1209-1227.
This includes both the original analysis and an alternative that uses a common weight matrix for all the restricted models.

This procedure calculates Clark and McCracken's MSE-t, MSE-F, ENC-t, and ENC-F forecast performance tests. (Clark and McCracken, 2001 and McCracken, 2004). Additional documentation is available in the file CLARKFORETESTDOC.PDF

Does a classical decomposition of a series into trend-cycle, seasonal and irregular components. The seasonals are assumed to be constant across the data range.

Procedures for computing various measures of persistence proposed by Cochrane in his 1988 JPE article.

Procedures for computing various measures of persistence proposed by Cochrane(1988), "How Big is the Random Walk in GNP?", JPE, vol 96, 893-920.

Computes Phillips-Ouliaris multivariate cointegration statistic.

Demonstrates several types of tests for cointegration. 1. Dickey-Fuller (using DFUNIT.SRC) test on a (hypothesized) cointegrating vector. 2. Engle-Granger test (using EGTEST.SRC) with an estimated cointegrating vector. 3. Johansen LR test (using JOHMLE.SRC)

Example of conditional forecasting and simulation in a VAR

Updated version of the conditional forecasting procedure that has been included with RATS for many years (see Section 10.14 of the RATS 6 User's Guide). This version allows more general linear constraints on the values of future endogenous variables.

Demonstrates a variety of tests for parameter constancy in a linear regression. 1. Does a Hansen/Nyblom fluctuations test, using the STABTEST.SRC procedure. 2. Does a Chow predictive test 3. A CUSUM test using the RLS instruction. 4. Sequential F-tests displayed graphically.

Estimates a consumer demand model using non-linear systems estimation (NLSYSTEM instruction). Demonstrates the use of named PARMSETS created with NONLIN, "adding" them to impose restrictions on the model's parameters.

Demonstrates use of a copula as an alternative to a multivariate GARCH model

Computes the non-parametric test statistic from Corrado (1989), "A Non-Parametric Test for Abnormal Security-Price Performance in Event Studies. Journal of Financial Economics, vol. 23, 385-395.

Computes a correlation integral for a series.

CrossCorr computes and graphs cross correlations of two series, presenting the information as a 2x2 matrix of graphs, with the autocorrelations of the two series, lag and lead cross correlations in separate graphs. (File was renamed from CROSSCOR.SRC in July, 2005)

Computes and optionally graphs the estimated coherence, phase and gain of a pair of series.

Symmetrizes the complex series cs

Demonstrates use of the CUMPDGM.SRC procedure for the Durbin cumulated periodogram test for white noise.

Performs a Durbin's Cumulated Periodogram for serial correlation. Durbin(1969), "Tests for Serial Correlation in Regression Analysis Based on the Periodogram of Least Squares Residuals", Biometrika, 1-16.

Replication file for Cushman and Zha(1997), "Identifying monetary policy in a small open economy under flexible exchange rates," Journal of Monetary Economics, vol 39, no 3, 433-448.
This estimates a structural near-VAR, with one block of variables taken as exogenous.

Performs CUSUM and CUSUMQ tests of the input series, which should be a series of recursive residuals.
Brown, Durbin & Evans(1975), "Techniques for Testing the Constancy of Regression Relationships over Time", JRSS-B, vol 37, 149-192.
Edgerton & Wells(1994), "Critical Values for the Cusumsq Statistic in Medium and Large Sized Samples," Oxford Bulletin of Economics and Statistics, vol. 56, no 3, 355-365.
Zeileis(2004), "Alternative Boundaries for CUSUM tests", Statistical Papers, vol 45, no 1, 123-131.

Demonstrates the use of the CVMODEL instruction for estimating structural VAR's.

Performs the special case of the Nyblom test for the case of a complete covariance matrix. Note that while this should have the correct asymptotic distribution under the null (under fairly typical assumptions in addition to stability), the argument for it being UMP against an alternative of martingale behavior won't hold because the covariance matrix is constrained to be positive definite.

This is a generalization of 0gDensity to complex matrices for use in multivariate Whittle likelihood estimation. It computes the frequency-by-frequency log likelihood. Use %CXLogDensityCV for the concentrated Whittle log likelihood.

This is a generalization of 0gDensityCV to complex matrices for use in multivariate Whittle likelihood estimation. This computes the concentrated log likelihood. Use %CXLogDensity for the frequency-by-frequency log likelihood.

Replication of Wouter J. den Haan(2000), "The comovement between output and prices," Journal of Monetary Economics, vol 46, no 1, 3-30.

Replication file for Dennis(2007), "Optimal Policy in Rational Expectations Models: New Solution Algorithms", Macroeconomic Dynamics, vol 11, 31-55.
Demonstrates use of the DSGEControl procedure.

A procedure for computing the non-parametric density function of a series using a standard normal kernel. Superseded by the introduction of the built-in DENSITY command in RATS Version 5.0, but useful for users with older versions.

Implements the proportional Denton method of benchmarking, distributing the sum from a low frequency series based upon the period-to-period rates of change of a (single) higher frequency series. This is designed to be used when the two series are closely related, with the more frequent series being a noisier measure of the less frequent one. The newer procedure disaggregate.src does the same calculation and more.

Procedure files for implementing the approximating filter for MS-GARCH models from Dueker(1997), "Markov Switching in GARCH Processes and Mean-Reverting Stock-Market Volatility," J of Business & Economic Statistics, vol. 15, no 1, 26-34.
The use of these is explained in the RATS "ARCH/GARCH and Volatility" e-course.

Procedure for computing (augmented) Dickey-Fuller Unit Root tests. Dickey and Fuller(1979), "Distribution of the Estimators for Time Series Regressions with a Unit Root", J.A.S.A., 427-431.

Replication files for Diebold and Yilmaz(2009), "Measuring Financial Asset Return and Volatility Spillovers, with Application to Global Equity Markets," Economic Journal, vol. 119, no. 534, 158-171.
This includes examples for the author's later "generalized spillover" measures.

Replication file for Diebold and Yilmaz(2012), "Better to give than to receive: Predictive directional measurement of volatility spillovers," International Journal of Forecasting, vol. 28, no 1, 57-66.

Computes DiGamma and TriGamma functions, which are the first and second derivatives of the log gamma function. Built-in functions for computing these (0IGAMMA and %TRIGAMMA) are now available in RATS beginning with version 6.02.

Does a temporal disaggregation of a series, changing the frequency to a higher one while maintaining the sum, average or final value over each period of the levels. This can handle a variety of models both for the noise term and linear, log-linear and multiplicative relationships. It does the work previously handled by the CHOWLIN, INTERPOL and DISTRIB procedures.

DISTRIB computes a distribution of a series, changing the frequency to a higher one while maintaining the sum each original period, e.g. producing a monthly "GNP" estimate from quarterly GNP. The newer procedure DISAGGREGATE.SRC can do the same calculation and quite a bit more.

Demonstrates various techniques for estimating distributed lags.
RATS Version 8, User's Guide, Example 2.5

Computes a Divisia index.

Estimation of a state-space model with a common growth cycle

Example of Kalman smoothing in a simple state-space model.

Example of Kalman filtering and out-of-sample forecasting in a simple state-space model.

Example of unconditional simulation in a simple state-space model.

Example of conditional simulation for a simple state-space model.

Estimates a linear regression by GLS, where the error process is described by a state space model, like those used by the RATS instruction DLM.

Computes and graphs impulse responses and error decompositions for the states in a state space model. The transition equation takes the form Y(t)=AY(t-1)+Fw(t) where F is an nxp matrix, where p is the number of fundamental shocks. The covariance matrix of w is assumed to be the identity.

Example of the use of the @DLMIRF procedure for doing impulse responses in a state-space model.

Computes Diebold-Mariano Forecast Comparison test, or, optionally, the Modified Diebold-Mariano test.
Diebold and Mariano(1995), "Comparing Predictive Accuracy," JBES, v.13, no.3, 253-63.
Harvey, Leybourne, Newbold(1997), "Testing the equality of prediction mean squared errors", International Journal of Forecasting, vol 13, pp 289-291.

Implements Stock and Watson's Dynamic OLS estimation of Cointegrating Vectors. We recommend that you download DOLS.ZIP a PK-Zip archive which includes the procedure, a read-me file, an example program and two datafiles. If you just want the procedure file, download DOLS.SRC.

Replication file for Diebold, Rudebusch & Aruoba (2006), "The macroeconomy and the yield curve: a dynamic latent factor approach," Journal of Econometrics, vol. 131(1-2), pages 309-338.

Procedure for doing optimal control on a linear rational expectations model. Algorithm from Dennis(2007), "Optimal Policy in Rational Expectations Models: New Solution Algorithms", Macroeconomic Dynamics, vol 11, 31-55.

Example of computation of the historical decomposition for a DSGE model.

Example of solution of a non-linear DSGE model
This uses the model from King, Plosser and Rebelo, but with the date conventions used by Uhlig and Sims, rather than KPR's.

Computes the solution to a linear rational expectations model input in the canonical form given in Sims, "Solving Linear Rational Expectations Models", Computational Economics, October, 2002.

Replication file for Dueker(1997), "Markov Switching in GARCH Processes and Mean-Reverting Stock-Market Volatility," J of Business & Economic Statistics, vol. 15, no 1, 26-34.
This estimates a variety of Markov Switching GARCH models.

A rough implementation of Dueker (2005), "Dynamic Forecasts of Qualitative Variables: A Qual VAR Model of U.S. Recessions", JBES, vol 23, no 1, 96-104, which estimates a VAR with a "probit" equation using Gibbs sampling.
(The data set has a version of the FedFunds rate from a different source, and uses the GDP deflator rather than the CPI)

Uses the Durbin-Levinson recursion to estimate the coefficients in a sequence of autoregressive representations for a stationary series.

Performs the calculations from Granger and Uhlig(1990), "Reasonable Extreme Bounds Analysis", J. of Econometrics, vol. 44, 159-170.

Demonstrates estimation of a vector error correction model. This uses the JOHMLE procedure to estimate the cointegrating vector, then imposes the error correction form using the ECT instruction.

Example of bootstrapping an EGARCH model (for forecasting variance out-of-sample)

Example of simulation of an EGARCH model (for forecasting variance out-of-sample)

Computes 'exact' critical values for the Dickey-Fuller and the Engle-Granger cointegration tests from the response surface regression in MacKinnon (1991).

Computes an Engle-Granger test for cointegration. The (approximate) critical values for t-test form are from MacKinnon, "Critical Values for Cointegration Tests", Long-Run Economic Relationships, R.F. Engle and C.W.J. Granger, eds, London, Oxford, 1991, pp 267-276. Use the related procedure EGTESTRESIDS if you already have the residuals.

Computes an Engle-Granger test for cointegration using the residuals from a previous first-stage regression, producing MacKinnon critical values. Use the related procedure EGTEST instead if you need to do the first stage regression as well.

Model specification for Erceg, Henderson & Levin(2000), "Optimal monetary policy with staggered wage and price contracts," Journal of Monetary Economics, vol. 46, no 2, 281-313.

Replicates results from Ehrmann, Ellison, Valla (2003), "Regime-dependent impulse response functions in a Markov-switching vector autoregression model", Economics Letters, Vol. 78, pp. 295-299.
The data set is a reconstruction rather than the authors' original data set.

Replication file for Elder and Serletis(2010), "Oil Price Uncertainty", Journal of Money, Credit and Banking, Vol. 42, No. 6, 1137-1159.

Computes the empirical likelihood for a set of moment conditions.

Replication file for Enders and Siklos(2001), "Cointegration and Threshold Adjustment," JBES, vol. 19, no. 2, 166-76.

Procedure for the threshold unit root tests in
Enders and Granger(1998), "Unit-Root Tests and Asymmetric Adjustment with an Example Using the Term Structure of Interest Rates", JBES, vol 16, pp 304-11.

Replicates Enders and Granger(1998), "Unit-Root Tests and Asymmetric Adjustment with an Example Using the Term Structure of Interest Rates", JBES, vol 16, 304-11. The technical details of the program are described on pages 128 - 132 of Enders' "RATS Progamming Manual".

Does various types of the unit root regressions with threshold breaks on the residuals from an Engle-Granger cointegrating regression.
Enders and Siklos(2001), "Cointegration and Threshold Adjustment," JBES, vol. 19, no. 2, 166-76.

Creates from the ARMA model given by an equation the theoretical autocovariance function

This implements the DFGLS, PT, DFGLSu and QT tests for unit roots due to Elliott, Rothenberg and Stock(1996), "Efficient Tests for an Autoregressive Unit Root", Econometrica, vol 64, no. 4, 813-836 and Elliott(1999), "Efficient Tests for a Unit Root When the Initial Observation is Drawn from its Unconditional Distribution", International Economic Review, vol 40, 767-783.

Computes the exact (limit) inverse of A + kB as k-->infinity, for p.s.d. symmetric matrices A and B. This is a matrix of the form C + k^-1 D.
This was introduced by Koopman(1997), "Exact Initial Kalman Filtering and Smoothing for Nonstationary Time Series Models," JASA, vol 92, pp 1630-1638. However, the calculation method has been simplified.

Introductory example #5 (linear regression with hypothesis tests).

Introductory example #4 (basic forecasting)

Introductory example #1 (display instruction)

Introductory example #6 (non-linear least squares)

Introductory example #3 (linear regression)

Introductory example #2 (data handling

Demonstrates forecasting using exponential smoothing with non-seasonal data. Uses the instruction ESMOOTH.

Demonstrates forecasting using exponential smoothing with seasonal data. Uses the instruction ESMOOTH.

Replication for Fabiani & Mestre(2004), "A system approach for measuring the euro area NAIRU," Empirical Economics, vol. 29, no. 2, pages 311-341.
Demonstrates state-space model with multiple observables.

Replication of Faust(1998), "The Robustness of Identified VAR Conclusions About Money", Carnegie-Rochester Conference Series on Public Policy, vol 49, 207-244.
This examines the maximal extent that GDP can be explained by a monetary shock in a VAR.

Replication file for Faust and Leeper(1997), "When Do Long-Run Identifying Restrictions Give Reliable Results", JBES, vol 15, no 3, pp 345-353.
Demonstrates use of the Blanchard-Quah factorization, and the procedures MCGraphIRF, MCFEVDTable and StructResids.

Fractional Integration Filter procedure. This is now (v 6.10) available in RATS with the RATS DIFF instruction.

Example provided by Kim and Nelson of a Markov Switching model with time-varying transition probabilities (TVTP).
Based upon Filardo(1994), "Business Cycle Phases and Their Transitional Dynamics", JBES, vol 12, no 3, 299-308 with a different data set.

Computes the Nyblom fluctuations test on a set of series of scores. Computes both individual and joint test statistics. Nyblom(1989), "Testing for Constancy of Parameters Over Time", JASA, vol 84, 223-230.

Estimates a cointegrating relation among the listed variables using fully modified least squares. Minor update in April, 2007 to fix issue with range used on final regression.
Phillips and Hansen(1990), "Statistical Inference in Instrumental Variables Regression with I(1) Processes", Review of Economic Studies, vol 57, 99-125.
Hansen, Bruce (1992), "Efficient Estimation and Testing of Cointegrating Vectors in the Presence of Deterministic Trends", Journal of Econometrics, vol 53, 87-121.

Fully modified estimation of cointegrating regressions.

Computes a factorization of a covariance matrix which includes (scales of) specified columns in the factorization, or, optionally, a scale of specified rows in the inverse of the factorization. This allows you to either force an orthogonalized component to hit the variables in a specific pattern (done by setting a column of the factorization), or to force that an orthogonalized component be formed from a particular linear combination of innovations (forces a row in the inverse).

Computes a wider range of fractiles than the standard STATISTICS(fract) command in RATS.

Demonstrates estimation of a fractionally differenced ARMA model (ARFIMA) using the Whittle likelihood.

Demonstrates frequency domain deseasonalization as described in Sims (reference??)

Example of calculation of Fry-Pagan (2011) median target estimator.
"Sign Restrictions in Structural Vector Autoregressions: A Critical Review," Journal of Economic Literature, vol. 49, no. 4, 938-60.

Computes and optionally graphs the estimated gain and phase of a pair of series.

Replication file for Jordi Gali(1999), "Technology, Employment and the Business Cycle: Do Technology Shocks Explain Aggregate Fluctuations", American Economic Review, vol. 89, pp. 249-271.

Replication file for Jordi Gali(1992), "How Well Does the IS-LM Model Fit Postwar U.S. Data", QJE, vol 107, no. 2, pp 709-738. Demonstrates VAR's with short and long run restrictions, particularly the instruction CVMODEL and the ShortandLong procedure.

Returns a 2-vector with the two parameters for a Gamma distribution with the given mean and standard deviation. Note that the RATS functions 0ggammadensity(x,a,b) and %rangamma(a) use a parameterization in terms of shape and scale, not degrees of freedom and mean. This returns the a and b appropriate for use with those functions.

A menu-driven procedure for estimating univariate ARCH, GARCH, and related models with varying lag lengths. By Norman Morin. Version 6, posted on November 17, 1999, adds support for MA terms in the mean equation, and the procedure code itself is now significantly shorter than before. Download GARCH.SRC (which includes non-negativity constraints) or GARCHN.SRC (identical, but omits non-negativity constraints). Note: RATS now has a built-in GARCH instruction for handling most ARCH and GARCH models, so this procedure is largely obsolete.

Example of rolling estimates for a GARCH model (for doing VaR calculations).

Example of bootstrapping with a univariate GARCH model, for calculating Value At Risk.
RATS Version 8, User's Guide, Example 16.6

Example of forecasting variance of a DCC GARCH Model

Example of estimation of a multivariate GARCH with DECO estimator.
Method (not example) from Engle and Kelly(2011), "Dynamic Equicorrelation", Journal of Business & Economic Statistics, vol 30, no 2, pp 212-228.

Example of use of fluctuation test for GARCH model

GARCHFore computes out-of-sample variance forecasts for a standard univariate GARCH model, estimated using the GARCH instruction.

Analysis of a univariate GARCH model using random walk Metropolis-Hastings.
RATS Version 8, User's Guide, Example 16.5

Example of variance GIRF from multivariate GARCH model.

Example of Monte Carlo integration on a GARCH model using importance sampling.
RATS Version 8, User's Guide, Example 16.3

An example of univariate GARCH-M with dummy shift in the "M" effect and variance.
Based upon a model from Sun and Tong(2010), "Risk and the January effect", Journal of Banking and Finance, vol 34, no 5, 965-974.

Multivariate GARCH examples. This program was originally written by Rob Trevor and discussed in May, 1994 issue of the RATSletter. It has since been updated by Estima to demonstrate newer aspects of the RATS code, and to include Engle's Dynamic Conditional Correlation (DCC) model. Note: RATS now has a built-in GARCH instruction for handling most ARCH and GARCH models, so the examples in this file are generally obsolete.

Example of bootstrapping for MV GARCH models. This estimates quantiles on bootstrapped 10-day returns using two different MV GARCH models.

Demonstrates multivariate GARCH with two-step DCC estimator.
Engle(2002), "Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models," JBES, vol 20, no 3, 339-350.

Example of Gibbs sampling (using Independence Chain Metropolis) with DCC model

Example of estimation of a multivariate GARCH model using MAXIMIZE.

Example of simulation of a multivariate (DVECH) GARCH process

Same as GARCH.SRC, but without non-negativity constraints.

Estimation of univariate GARCH model with semi-parametric handling of the density of the error process. Partially described in RATS Version 8 User's Guide, section 13.1

Univariate GARCH examples. This program was originally written by Rob Trevor and discussed in May, 1994 issue of the RATSletter. Note: RATS now has a built-in GARCH instruction for handling most ARCH and GARCH models, so the examples in this file are generally obsolete.

Estimates univariate GARCH models with non-standard density functions using GARCH with the DENSITY option.

Examples of estimates of univariate GARCH models using MAXIMIZE.

Example of computing the Volatility Impulse Response (VIRF) for a GARCH model which doesn't have a "VECH" form, but only models the variances.

GaussHermite returns the points at which f(x) needs to be evaluated and the corresponding weights for Gauss-Hermite numerical integration.

Demonstrates the GCONTOUR instruction for creating contour graphs.

Generates draws for a generalized error distribution by acceptance-rejection method. Revised in April, 2006, to add SHAPE option.

Systematically generates all combinations of k integers from the set {1,...,n}. All this procedure does is to generate them as the vector of integers "pick". If you need to do something more than this, insert the necessary code at the "disp pick" instruction.

Demonstrates Gibbs sampling for a linear regression model with a Normal-gamma prior. Uses DENSITY instruction to generate estimated density function for parameters.

Demonstrates Gibbs sampling estimation of a dynamic "probit" model, where the latent index follows an AR(1) equation.

This is an example of the use of Gibbs sampling applied to a VAR with a standard Minnesota prior. Different priors can be handled by changing the way that bprior and hprior (the mean and precision of the prior) are created.

Example of Gibbs sampling on a Bayesian VAR with a "Minnesota" prior using @BVARBuildPriorMN to create the matrices for the prior.

Uses Gibbs sampling to compute posterior distributions of the parameters of a VAR, along with forecasts. The model/estimator is the normal-diffuse of Kadiyala and Karlsson (1997), "Numerical Methods for Estimation and Inference in Bayesian VAR Models," Journal of Applied Econometrics, pp 99-132

Demonstrates generalized instrumental variables estimation. Uses the NLLS instruction with the INSTRUMENTS and OPTIMAL options.

Local to unity GLS detrending routine. Includes the ERS cases (no constant, constant, constant and trend) as well as the Perron and Rodriguez cases (constant and trend with a single break). This just does the detrending, not the actual unit root test(s).
Elliott, Rothenberg and Stock(1996), "Efficient Tests for an Autoregressive Unit Root", Econometrica, vol 64, no. 4, pp 813-836.
Perron and Rodriguez(2003), "GLS Detrending, Efficient Unit Root Tests and Structural Change", Journal of Econometrics, vol 115, pp 1-27.

Does an automatic fit of a multiplicative seasonal ARMA model to a series following the procedure described in Gomez and Maravall, "Automatic Modeling Methods for Univariate Series", in Peña, Tiao and Tsay, eds., "A Course in Time Series Analysis", New York: Wiley, 2001.

Performs the Granger-Newbold forecast comparison test.
Granger and Newbold(1973), "Some Comments on the Evaluation of Economic Forecasts", Applied Economics, vol. 5, 35-47.

Replication file for González-Rivera(1998), "Smooth-Transition GARCH Models," Studies in Nonlinear Dynamics & Econometrics, vol. 3, no 2, 1-20.

Replication of the interest rates example in Gonzalo and Granger(1995), "Estimation of common long memory components in cointegrated systems", JBES, vol 13, pp 27-36. Demonstrates the procedure johmle.src and the function %PERP.

Estimates fractional difference power for series using the frequency domain regression techniques of Geweke and Porter-Hudak(1983), "The Estimation and Application of Long Memory Time Series Models", J. of Time Series Analysis, pp 221-238.

Bootstrapped p-values for Granger causality test.
Alternative to CAUSAL.RPF
RATS Version 8, User's Guide, Example 3.2

Demonstrates creating of a boxplot using the GBOX instruction.

Demonstrates graphing forecasts (time series defined over different intervals). Uses the ESMOOTH instruction for forecasts.

Demonstrates use of the SCATTER instruction with STYLE=LINES for graphing a general function.

Demonstrates "high-low-close" graphs, done using the GRAPH instruction with the option STYLE=VERTICAL.

Demonstrates the various labels on a GRAPH instruction.

GraphMatrix graphs a rectangular array of series on separate graphs The user can control the maximum number of graphs which can be placed either horizontally or vertically; if needed, the procedure will create as many pages as necessary in order to display the graphs.

Demonstrates "overlay" graphs (graphs with separate scales for the left and right axes).

Replicates work on Markov-switching GARCH models from Gray(1996), "Modeling the conditional distribution of interest rates as a regime-switching process", J. of Financial Economics, vol 42, pp 27-62. Makes extensive use of the functions on markov.src.

Implements the Gregory-Hansen cointegration test. The cointegrating regression is allowed to have a trend or not, and can have either a break in the intercept only or a break in all coefficients.
Gregory and Hansen(1996), "Residual-based Tests for Cointegration in Models with Regime Shifts", Journal of Econometrics, vol 70, 99-126.
Gregory and Hansen(1996), "Tests for Cointegration in Models with Regime and Trend Shifts," Oxford Bulletin of Economics and Statistics, vol. 58, no 3, pp 555-560.
(Renamed from GREGHANS.SRC in July, 2005).

Replication of example from Gregory and Hansen(1996), "Residual-based Tests for Cointegration in Models with Regime Shifts", Journal of Econometrics, vol 70, 99-126.

Creates a series with an evenly spaced grid based either upon range and number of points, or range and interval width.

Replication file for Grier, Henry, Olekalns & Shields(2004), "The asymmetric effects of uncertainty on inflation and output growth," Journal of Applied Econometrics, vol 19, no. 5, 551-565.
This requires RATS 8.2 or later because of the use of the SIGNS option on the GARCH instruction to allow for asymmetric effects with different signs for output and inflation.

Implements Hadri tests for unit roots in panel data. This has stationarity as the null, and rejects if the detrended or de-meaned data are too persistent.
Hadri(2000), "Testing for stationarity in heterogeneous panel data", The Econometrics Journal, vol 3, no 2, 148-161.

Replication file for Hafner and Herwartz(2006), "Volatility impulse responses for multivariate GARCH models: An exchange rate illustration", Journal of International Money and Finance, vol 25, no 5, 719-740.
Note: the data set is a reconstruction.

Generates an n-vector of the Halton sequence for the given base (which should be a prime number). Halton sequences are deterministic "low-discrepancy" sequences which fill the unit cube relatively uniformly (with a different base for each dimension) allowing numerical integrals to be estimated more accurately for a given number of function evaluations than would be possible with truly random numbers.

Updated version of the HAMILTON.PRG switching model example discussed in the RATS 5.0 User's Guide. This version includes an EM algorithm written by Andrew Leach that provides better initial conditions, allow the model to converge to a solution when using the entire sample data set.

Replication file for Hamilton and Susmel(1994), "Autoregressive Conditional Heteroskedasticity and Changes in Regime," Journal of Econometrics, vol 64, pp 307-333.
This does Markov Switching ARCH models.

Demonstrates Hannan Efficient GLS estimation for a distributed lag.

Computes estimates for an ARMA model using the Hannan-Rissanen algorithm, which runs a LS regression on lags of the dependent variable and residuals from a long preliminary AR.
Hannan and Rissanen, "Recursive estimation of mixed autoregressive-moving average order", Biometrika, 1991, vol 69, pp 81-94.

Demonstrates use of LINREG with INSTRUMENTS for testing overidentifying restrictions.

Replication of Bruce Hansen (1994), "Autoregressive Conditional Density Estimation", International Economic Review, vol 35, no. 3, pp 705-730. This estimates GARCH models with student t errors with time-varying degrees of freedom, and introduces the skew-t density.

Replication file for the examples of Hansen, Bruce E.(1997), "Approximate Asymptotic P-Values for Structural Change Tests", Journal of Business and Economic Statistics, vol 15, no. 1, pp 60-67.
Uses APBREAKTEST procedure.

Replication file from Bruce Hansen(1999), "Threshold effects in non dynamic panels: estimation, testing and inference", Journal of Econometrics, vol 93, pp 345-368.
Does threshold effects on a fixed effects estimator in panel data. Demonstrates use of PANELTHRESH procedure.

Replication for Hansen(2000), "Testing for Structural Change in Conditional Models", Journal of Econometrics, vol. 97, no. 1, pages 93-115.
Demonstrates test for a structural break in a linear regression model with fixed regressor bootstrap.

Replication program for a SETAR model on GNP from Bruce Hansen(1996), "Inference When a Nuisance Parameter is Not Identified Under the Null Hypothesis", Econometrica, vol 64, no. 2, pp 413-430. Demonstrates use of tar.src procedure.

Replication file for Hansen and Seo(2002), "Testing for two-regime threshold cointegration in vector error-correction models", Journal of Econometrics, vol 110, pp 293-318.

Replication file for Harvey, Ruiz and Shephard(1994), "Multivariate Stochastic Variance Models", Review of Economic Studies, vol 61, no. 2, pp 247-264. Includes programs for univariate and multivariate models.

Example of the calculation of decomposition of long-run variance using the techniques from Hasbrouck(1995) "One security, many markets: determining the contribution to price discovery", Journal of Finance, vol 50, no 4, pp 1175-1199.

Demonstrates direct calculation of a Hausman specification test statistic for 2SLS vs 3SLS.

An implementation of the "HEGY" (Hylleberg, Engle, Granger, and Yoo, 1990) seasonal unit root tests for quarterly time series. Options allow you to specify the number of lags, and choose between models with and without intercept, seasonal dummies, and/or trend. You can also have it test all of these models at once.

Demonstrates LINREG with the SPREAD option for estimating regression with heteroscedastic errors. Includes examples with both variances being known and being estimated.

Demonstrates Goldfeld-Quandt, Breusch-Pagan and Harvey tests for heteroscedasticity.

Estimates the tail index for a distribution, using Hill's method.
Hill(1975), "A Simple General Approach to Inference About the Tail of a Distribution," Annals of Statistics, vol 3, 1163-1174.

Performs the Hinich bispectrum test for linearity and Gaussianity. Hinich and Patterson (1989), "Evidence of nonlinearity in the trade-by-trade stock market return generating process", in Economic Complexity: Chaos, Sunspots, Bubbles and Nonlinearity, Barnett, Geweke and Shell, eds. Cambridge University Press.

A modified version of the older HIST.SRC (see below). This version displays bin ranges and counts.

New Histogram procedure, using the DENSITY command introduced in RATS 5.0. This includes one command that requires 5.04 or later--this command can be removed if you have an older version (see comment line in the file). (Renamed from HISTOGRM.SRC).

Demonstrates calculation and graphing of the historical decomposition of the data using a VAR.

HISTSCAT generates a histogram plot using SCATTER with the XLABELS option. It can't do bar graphs, but it does provide proper x-axis labeling.

HJBounds computes (and optionally graphs) the Hansen-Jagannathan bounds for a set of returns, as a function of the unobserved mean of a riskfree asset.
Hansen and Jagannathan(1991), "Implications of Security Market Data for Models of Dynamic Economics", JPE, vol 99, 225-262.

Example of techniques for estimating a VAR with panel data (with large N-small T data set) from Holtz-Eakin, Newey and Rosen(1988), "Estimating Vector Autoregressions with Panel Data," Econometrica, vol. 56, no 6, pp 1371-95.
The data set is from the same source as Holtz-Eakin, Newey and Rosen(1989), "The Revenues-Expenditures Nexus: Evidence from Local Government Data," International Economic Review, vol 30, no 2, 415-429, but looks at a different model.

Demonstrates various ways of computing the Hodrick-Prescott filter using RATS. The standard HP filter is built into the FILTER instruction (beginning with RATS 6.20), but it can also be analyzed using the instruction DLM.

Executes a Hodrick-Prescott Filter (Hodrick, R. & Prescott, E., "Post-War U.S. Business Cycles: An Empirical Investigation", Carnegie-Mellon working paper, 1980. NOTE: The FILTER instruction in RATS can now compute the HP filter directly.

Implementation of Harris-Tzavalis test for unit roots in panel data. This has a null of a unit root, with asymptotics assuming large N, fixed T.
Harris and Tzavalis(1999), "Inference for unit roots in dynamic panels where the time dimension is fixed". Journal of Econometrics, vol 91, pp 201–226.

Computes a Hurst exponent. Revised in March, 2007 to use a single dialog box for input, and for improved output.

Performs the Inclan-Tiao test for breaks in variance.
Inclan and Tiao(1994), "Use of Cumulative Sums of Squares for Retrospective Detection of Changes in Variance", JASA, vol 89, 913-923.

Imhof Procedure for computing P(u'Au < x) for a quadratic form in Normal(0,1) variables. This can be used for ratios of quadratic forms as well, since P((u'Au/u'Bu) < x)= P(u'(A-xB)u < 0). Use the functions %QFORMPDF or %QFORMDPDF with RATS versions 6.00 or later.

This procedure is used to correct the signs of columns in a factor of a covariance matrix when the signs themselves may not be identified by the process of computing the factor (such as with short- and long-run restrictions).

Demonstrates calculation and (various ways of) graphing impulse response functions. It also demonstrates with VARIRF.SRC procedure, which can do these types of graphics as well.

Replication file for Inclan and Tiao, "Use of Cumulative Sums of Squares for Retrospective Detection of Changes in Variance", JASA 1994, vol 89, pp 913-923. Data are included in the program file.

Demonstrates looping over graph instructions.

Demonstrates basic 2SLS estimation, using the instruction LINREG with the INSTRUMENTS option.

Interpolates a series, changing the frequency to a higher one while maintaining the last value in each period. The newer procedure DISAGGREGATE.SRC can do the same calculation and quite a bit more.

Demonstrates intervention analysis using the instruction BOXJENK with the INPUTS option.

Returns a 2-vector with the two parameters (degrees of freedom and scale, in that order) for an inverse chi-square distribution with the given mean and standard deviation.

Returns a 2-vector with the two parameters for an inverse Gamma distribution with the given mean and standard deviation. Note that the RATS functions 0ggammadensity(x,a,b) and %rangamma(a) use a parameterization in terms of shape and scale, not degrees of freedom and mean. This returns the a and b appropriate for use with those functions.

Implements panel unit root tests from Im, Pesaran and Shin(2003), "Testing for Unit Roots in Heterogeneous Panels", J. of Econometrics, vol 115,53-74.

Replication program for Ireland(2004), "A Method for Taking Models to the Data", Journal of Economic Dynamics and Control, vol 28, no. 6, 1205-1226.
Demonstrates use of DSGE instruction, combined with DLM, for estimating a dynamic model using maximum likelihood.

Builds up (one constraint at a time) a restriction matrix for a "B" form structural VAR.

Computes Johansen lambda tests for cointegrating rank, and the ML estimator for a single cointegrating vector. Updated in February, 2006, October, 2006, and February, 2007, to include additional options and improved output.

Replication file for Jorda(2005), "Estimation and Inference of Impulse Responses by Local Projections", American Economic Review, vol 95, no 1, 161-182.

Replication file for Jacquier, Polson and Rossi(1994), "Bayesian Analysis of Stochastic Volatility Models," Journal of Business and Economic Statistics, vol 12, no 4, pp. 371-89.
Includes both a Metropolis MCMC and a rejection method similar to that in Kim, Shephard and Chib(1998), "Stochastic Volatility: Likelihood Inference and Comparison with ARCH Models", Review of Economic Studies, vol 65, pp 361-93.

KERNEL.SRC computes and graphs a non-parametric estimate of the unconditional density of a single series. Other than the graphics, this can be done with with the RATS instruction DENSITY with any version 5.0 or later.

KERNREG.SRC and LOWESS.SRC are procedures for flexible fits of the form y = f(x). KERNREG does this by kernel regression (weighted sums of y values), while LOWESS uses locally weighted fits of first order polynomials. RATS v5 now includes the NPREG instruction.

KFEXACT.SRC performs the Kalman filtering technique described in "Exact Initial Kalman Filtering and Smoothing for Nonstationary Time Series Models" by Siem Jan Koopman, JASA, December 1997. This is now included in the RATS instruction DLM when you use the EXACT option.

Kalman filtering procedure, using general matrix inputs (updated so that you don't have to declare/dimension the XKK array ahead of time). This procedure has been superseded by the built-in DLM command introduced in RATS 5.0.

Replication of Kilian(1998), "Small-Sample Confidence Intervals for Impulse Response Functions", Review of Economics and Statistics, vol 80, no 2, 218-230.

Replication file for Kilian & Vigfusson(2011), "Are the responses of the U.S. economy asymmetric in energy price increases and decreases?", Quantitative Economics, vol. 2, no 3, 419-453.

Bootstrap-in-bootstrap analysis of a VAR with asymmetries.

Estimates Klein's Model I using 2SLS

Computes the test statistic for the Kolmogorov-Smirnov test of goodness of fit and displays the corresponding critical value at different significance levels. Frequently used to test for Normality, can also be used to test for Chi-squared, F, and t-distributions.

This is an example of the Gibbs sampling procedure for cointegrated models described in Koop, León-González and Strachan(2010), "Efficient Posterior Simulation for Cointegrated Models with Priors on the Cointegration Space", Econometric Reviews, vol. 29, no. 2, 224-242.
This actually does just the flat prior, though adding the types of priors they describe isn't difficult.

Replication file for the multivariate EGARCH model with mean and asymmetric volatility spillovers from Koutmos(1996), "Modeling the Dynamic Interdependence of Major European Stock Markets", Journal of Business Finance and Accounting, vol 23, 975-988.

Does the KPSS stationarity test procedure.
Kwiatowski, Phillips, Schmidt & Shin(1992), "Testing the Null Hypothesis of Stationarity Against the Alternative of a Unit Root: How Sure Are We That Economic Time Series Have a Unit Root?", J. of Econometrics, vol 54, 159-178.

Replicates results from King, Plosser, Stock and Watson(1991), "Stochastic Trends and Economic Fluctuations", American Economic Review, vol. 81, pp. 819-840. Demonstrates the procedures swdols.src, swtrends.src, johmle.src, and forcedfactor.src

Replication files for Krolzig's "International Business Cycles: Regime Shifts in the Stochastic Process of Economic Growth", Oxford University Discussion Paper.

Uses the rejection method to draw from the posterior formed by multiplying a Normal by a log inverse gamma. This comes up in the analysis of the stochastic volatility model. This is a refinement of the proposal in:
Kim, Shephard and Chib(1998), "Stochastic Volatility: Likelihood Inference and Comparison with ARCH Models", Review of Economic Studies, vol 65, pp 361-93

KSMOOTH.SRC performs Kalman smoothing, using general matrix inputs. See also the example program KSMOOTH.PRG. Note that this procedure has been superseded by the built-in DLM instruction introduced in RATS 5.0.

Example of L1 filtering (to extract a trend).
Kim, Koh, Boyd, Gorinevsky(2009), "L1 Filtering", SIAM Review, vol 51, pp 339-360.

Produces a table of the inverted roots of the input lag polynomial, showing the modulus and (for complex roots) the period.

Automatically computes various lag-length tests for a single series, including AIC, BIC, Ljung-Box, and more.

Replication file for Lanne and Lutkepohl(2008), "Identifying Monetary Policy Shocks via Changes in Volatility", JMCB, vol 40, no 6, 1131-1149.
Demonstrates identification of a structural VAR using volatility regimes.

Replication file for Laubach and Williams(2003), "Measuring the Natural Rate of Interest", Review of Economics and Statistics, vol. 85, no 4, 1063-1070.
Estimates several state-space models with multiple observables with a combination of regression equations and latent variables.

Replication file for Lebo and Box-Steffensmeier(2008), "Dynamic Conditional Correlations in Political Science", American Journal of Political Science, vol 53, no 2, 688-704.
Does multivariate GARCH models with DCC applied to political science data.

Replication for T. H. Lee(1994), "Spread and volatility in spot and forward exchange rates," Journal of International Money and Finance, vol. 13, no 3, 375-383.

This estimates a set of VECM-GARCH-X models on spot and forward exchanges rates for several countries.

Implements tests from Levin, Lin and Chu(2002), "Unit root tests in panel data: Asymptotic and finite-sample properties", Journal of Econometrics, vol 108, no 1, 1–24.

LIML does limited information maximum likelihood estimation of a linear regression equation. Before using this, you should set the instrument set using INSTRUMENTS.

Example of the use of the LIST aggregator

Creates the "A", "C" and "F" matrices for a local level or local trend DLM with shocks to the trend rate or level or both.

Provides rough estimates of the component variances for the irregular and trend rate variances for a local level or local trend model, possibly in the presence of seasonals.

Performs a "local smoothing regression". Used in one of the example programs for the Makridakis, Wheelwright, and Hyndman textbook (see Textbook Examples)

Computes the log density function for a multivariate skew-t distribution.
Bauwens & Laurent(2005), "A New Class of Multivariate Skew Densities, With Application to Generalized Autoregressive Conditional Heteroscedasticity Models," JBES, vol 23, pp 346-354.

Returns a 2-vector with the parameters for the underlying Normal to achieve the given mean and standard deviation for a log normal.

0gSkewGEDDensity(x,h,lambda,k) returns the log density at x of the univariate skewed GED with mean zero and variance h.

From Panayiotis(2015), "Skewed Generalized Error Distribution of Financial Assets and Option Pricing," Multinational Finance Journal, vol. 19(4), pages 223-266.

0gSkewGEDGARCH(h,x) returns the log density at x of the univariate skewed GED with mean zero and variance h. This is for use with the RATS GARCH instruction with the DENSITY option so the skewness and kurtosis control parameters are put into global variables.

From Panayiotis(2015), "Skewed Generalized Error Distribution of Financial Assets and Option Pricing," Multinational Finance Journal, vol. 19(4), pages 223-266.

Function returning the log of the normalized (to unit variance and zero mean) skew-t density from Hansen(1994), "Autoregressive Conditional Density Estimation", International Economic Review, vol 35, no. 3, 705-730.

0gSkewTGARCH(h,z) returns the log of the normalized (to variance h and zero mean) skew t density. This is for use with the RATS GARCH instruction with the DENSITY option, so the skewness and kurtosis control parameters are handled using global variables.

From Hansen (1994), "Autoregressive Conditional Density Estimation", International Economic Review, vol 35, no. 3, pp 705-730, the

Example of non-parametric regression techniques.
From the NIST Engineering Statistics Handbook

Does a flexible fit of y=f(x) using the lowess (locally weighted scatterplot smoothing). This is obsolete. Use the RATS instruction NPREG instead.

Implements the Lumsdaine-Papell unit root test, allowing for two breaks in the intercept, the trend or both at unknown locations.
Lumsdaine and Papell(1997), "Multiple trend breaks and the unit root hypothesis", Review of Economics and Statistics, vol 79, 212-218.

Computes a least squares dummy variable (i.e. fixed effects) estimator for a dynamic panel data model with correction for bias.
Reference: Kiviet(1995), "On bias, inconsistency, and efficiency of various estimators in dynamic panel data models," Journal of Econometrics, vol. 68, no. 1, 53-78.

Implementation of Lee-Strazicich LM unit root tests with one or two structural breaks. Optionally, it can do more than two breaks.
Lee and Strazicich(2004), "Minimum LM Unit Root Test with One Structural Break", Appalachian State University Working Paper
Lee and Strazicich(2003), "Minimum LM Unit Root Test with Two Structural Breaks", Review of Economics and Statistics, vol 85, no. 4, 1082-1089.

LTS.SRC implements the Least Trimmed Squares Regression method. This computes robust linear regressions in the presence of outliers. For details on this technique and the procedure, see LTS Information (HTML file), or download the same info in the file LTS.TXT).

Replication for the DSGE model from Lubik, Thomas A. & Schorfheide, Frank(2007), "Do central banks respond to exchange rate movements? A structural investigation," Journal of Monetary Economics, vol 54, no. 4, 1069-1087.
Includes a general solution of the model (for examining dynamics) and estimation with Canadian data.

@MAAutoLags computes an information criterion for various lags of MA processes using the innovations algorithm, finding the "optimal" choice under that criterion and (optionally) producing a table with the values for all lags.

Computes "exact" critical values for the Dickey-Fuller and the Engle-Granger cointegration tests from the response surface regressions in MacKinnon, J. (1991), "Critical Values for Cointegration Tests", Long-Run Economic Relationships, R.F. Engle and C.W.J. Granger, eds. London: Oxford University Press. (Renamed from MACKCVAL.SRC).

Performs a Mann-Whitney(-Wilcoxon) non-parametric test for whether the observations from the vectors X and Y are drawn from the same distribution.
Wilcoxon(1945), "Individual comparisons by ranking methods", Biometrics Bulletin, vol 1, pp 80–83.
Mann & Whitney(1947), "On a test of whether one of two random variables is stochastically larger than the other", Annals of Mathematical Statistics, vol 18, pp 50–60.

Replication for Mark and Sul(2003), "Cointegration Vector Estimation by Panel DOLS and Long-run Money Demand," Oxford Bulletin of Economics and Statistics, vol. 65, no. 5, 655-680.
Demonstrates DOLS for panel data.

MARKOV.PRG provides examples of Markov switching models using generated data. Based on December 1998 RATSLetter article. This has been superseded by the newer HAMILTON.PRG example in RATS Version 5.0, so these will only be of interest to users still running RATS 4.x, who can refer to the December 1998 newsletter for technical details.

Based upon Matheson and Stavrev(2013), "The Great Recession and the inflation puzzle," Economics Letters, vol. 120, no 3, pp 468-472 with a reconstructed (and extended) data set.
Estimates the model by a non-linear Kalman filter using unconstrained estimates, inequality constrained state estimates and state estimates constrained through use of functions that map to the proper range.

Contains two procedures: MatrixPeek and MatrixPoke. These extract information or put information into a rectangular array at specified locations. Since version 5.10, RATS now includes built-in functions %MATPEEK and %MATPOKE.

Example of the use of MAXIMIZE for a stochastic frontier model.
Adapted from Greene, Econometric Analysis, 6th Edition, example 16.9 from pp 541-2.

Version of KERNEL.SRC required by MODES.SRC.

Computes error bands for the forecast error variance decomposition for a VAR using using the information already computed. Note: this assumes that the shocks used are orthogonal and produce a complete factorization of the covariance matrix. It cannot be used with isolated shocks (from, for instance, sign restrictions).

Graphs error bands for impulse response functions using the information already computed by another procedure (such as MCVARDoDraws). This has many options controlling the general layout and content.

Performs a McLeod-Li test for 2nd order dependence. McLeod and Li(1993), "Diagnostic Checking of ARMA Time Series Models Using Squared Residual Autocorrelations", J. of Time Series Analysis, vol 4, pp 269-273.

Post processor for Markov Chain Monte Carlo statistics. Computes means, standard errors, numerical standard errors and CD measures for each component in an input SERIES[VECT] with the generated statistics.

Monte Carlo option pricing using antithetic acceleration.

Generates error bands for impulse response functions using the information already computed by another procedure (such as MCVARDoDraws). This includes the same calculations as is done by MCGraphIRF, but rather than graphing, creates three RECT[SERIES] which can be graphed or put into tables as you need.

Computes impulse response functions for a VAR using Monte Carlo simulation. This assumes the model is an OLS VAR estimated using the ESTIMATE instruction. This uses a Choleski factorization, though it can be modified fairly easily to do any just-identified factorization.

Similar to SWAMY.SRC, but does a simple weighted average of the coefficient vectors.

Displays a mean excess plot for a series.

Computes and optionally graphs a spectrum using Maximum Entropy Method.

MHEGY.SRC implements the monthly version of the "HEGY" (Hylleberg, Engle, Granger, and Yoo, 1990) seasonal unit root tests (use the HEGY.SRC version above for quarterly series). Note: Requires Norman Morin's LAGSELEC.SRC procedure file.

Replication for Michael, Nobay and Peel(1997), "Transactions Costs and Nonlinear Adjustment in Real Exchange Rates: An Empirical Investigation", J. of Political Economy, vol 105, no 4, pp 862-879.

Estimates an equation using the mixed estimation technique. See the RATS User's Guide for full details.

Example of a mixture model (not-Markovian) demonstrating the different estimation strategies.

MIXVAR computes estimates for a single equation using the mixed estimation procedure used by the RATS command ESTIMATE. It can be helpful when you have a variable you wish to forecast which does not fit into a full-blown VAR, because it has no predictive content for the other variables, or when you have a limited amount of data for one variable, so you need to use fewer coefficients in that equation than in others.

Replication file for Morley, Nelson & Zivot(2003), "Why Are the Beveridge-Nelson and Unobserved-Components Decompositions of GDP So Different?," The Review of Economics and Statistics, vol. 85(2), pages 235-243.

Defines a function "%ModelCompanion(model)" which returns the companion matrix for a model (could be a VAR, but doesn't have to be). Beginning with version 6.35, this function is built-in.

Function for using %ModelLagMatrix for RATS versions before 8.1. This returns the NxN matrix of coefficients for the dependent variables of the <> at lag <>.

MODES.SRC finds the critical windows and the number of modes of a given time series by KERNEL looping from an initially small window up to the largest obtained through selected increments and incremental values. The companion procedure BOOTS.SRC retrieves critical windows stored in a temp file, KERNEL boots the given time series for each of them, and tests for the null of m versus the alternative of m+1 modes by computing the ASL (Achieved Significance Level) statistics. Both require MBKERNEL.SRC, a stripped-down version of the KERNEL.SRC procedure. MBDEMO.PRG and GDPQ.RAT are sample program and data files. Or download MODES.ZIP which contains all of the above files.

Demonstrates Monte Carlo examination of the critical values of a test; in this case, for a test for ARCH.

Demonstrates Monte Carlo integration in a VAR with shock to an "exogenous" variable.

Example of MCMC analysis of a combination of a near VAR for the lag coefficients and a structural VAR for the covariance matrix.

Demonstrates an efficient method of drawing from the posterior distribution of a near-VAR by means of importance sampling for the covariance matrix.

Monte Carlo (importance sampling) integration for impulse responses in a structural VAR.
montesvar_mh.rpf is similar, but uses Metropolis-Hastings

Monte Carlo integration of a structural VAR using Random Walk Metropolis. (MONTEVAR.RPF is similar, but uses importance sampling).

Example of Monte Carlo Integration of Impulse Responses
Note that this is for illustration of the technique. In practice, you would use the @MONTEVAR procedure to do the draws.

This computes and graphs error bands for impulse response functions for a VAR using Monte Carlo simulation. This assumes the model is a symmetric VAR estimated using the ESTIMATE instruction. This uses a Choleski factorization, though it can be modified fairly easily to do any just-identified factorization.

Monte Carlo integration for a Vector Error Correction Model (with fixed cointegrating vector)

Replication file for Mountford and Uhlig (2009), "What are the Effects of Fiscal Policy Shocks?", Journal of Applied Econometrics (to appear).
Demonstrates analysis of impulse responses with sign constraints, and sign + zero constraints.

These functions and procedures are fairly generic calculations for EM estimation in Markov Switching models.

Procedure file for Markov Switching univariate linear regressions, with the either the full coefficient vector switching or part of the coefficient vector is switching and part fixed. Use the MSVARSetup procedures for switching autoregressions and VAR's, and MSSysRegression for other types of multivariate regressions.

These are fairly generic Markov Chain/Markov Switching model support functions. This takes the place of the markov.src file.

Procedure file for Markov switching multivariate linear regressions (same right hand side variables) with either the full coefficient vector switching or part of the coefficient vector is switching and part fixed. Use the MSVARSetup procedures for switching autoregressions and VAR's where only the mean (and possibly the variance) changes.

Model of Markov-switching variances.
Based upon Application 3 from pp 86-92 of Kim and Nelson, "State-space Models with Regime Switching"

Procedure file for setting up Markov switching VAR's.

This does multiple structural change analysis as described in Bai and Perron, but using a threshold variable other than time.
Bai and Perron(2003), "Computation and Analysis of Multiple Structural Change Models", J. of Applied Econometrics, pp 1-22.

Performs a multivariate LM test for ARCH effects in a set of series by regressing the crossproducts of the series (that is u(i,t) x u(j,t) for all combinations of i and j) on a constant and its lag(s) and testing the coefficients on the lags.

Computes a multivariate Beveridge-Nelson decomposition of a set of series via a vector autoregression.
Arino and Newbold(1998), "Computation of the Beveridge-Nelson Decomposition for Multivariate Economic Time Series", Economic Letters, vol 61, 37-42.

Forecasts a multi-variate GARCH model, estimated using the GARCH instruction. Minor revisions/corrections in May 2006.

Extracts a VECH representation out of the most recent GARCH instruction for use in forecasting or impulse response analysis.

@MVGARCHVarFore computes out-of-sample variance forecasts for several types of GARCH models which don't have a full VECH representation. As a result, they forecast only the variances (producing zeros for the off-diagonals).

This pulls the matrices for a variance recursion out of the output from a GARCH model. Do this immediately after the GARCH instruction and repeat the VARIANCES option.

Creates a Tiao-Box cross correlation matrix with +,- and . symbols for significant positive, negative and insignificant cross correlations respectively.
Tiao and Box(1981), "Modeling multiple time series with applications", JASA, vol 76, 802-816.

Computes a multivariate version of the Jarque-Bera test for normality. Note that there are more sophisticated versions of this (for instance, Doornik and Hansen, "An Omnibus Test for Univariate and Multivariate Normality"). This just transforms the input residual series to uncorrelated components (using an eigenbased factorization if not provided by the user) and sums up the univariate JB statistics from those.

Multivariate Kalman filtering procedure. Superseded by built-in DLM instruction introduced in RATS 5.0.

Computes the Hosking variant on the multivariate Q statistic.
Hosking(1981), "Equivalent Forms of the Multivariate Portmanteau Statistic", JRSS-B, vol 43, no. 2, 261-262.

Generates dummy variables for quarterly or monthly data based upon the NBER business cycle reference dates.
Source:
National Bureau of Economic Research, Inc.
1050 Massachusetts Avenue
Cambridge MA 02138
USA
www.nber.org/cycles.html

Estimates and does predictions using a neural network model. Demonstrates the instructions NNLEARN and NNTEST.

Demonstrates estimation using non-linear least squares with the instruction NLLS.

Example of dynamic forecasting with neural networks

Demonstrates several techniques for maximum likelihood estimation of a non-linear model.
RATS Version 8, User's Guide, Example from Section 4.13

Implements a multivariate test for normality, as described by Doornik and Hansen (1994), which was based on a proposal of Shenton and Bowman (1977). It can also perform univariate normality tests. The test is described in a working paper by Jurgen A. Doornik and Henrik Hansen, which is available as a PostScript file at: http://www.nuff.ox.ac.uk/Users/Doornik/.

Demonstrates nonparametric regression using the NPREG instruction. The application is semiparametric weighted least squares, estimating the schedastic function for weighted least squares using NPREG.

Estimates an observable index model from Sargent & Sims(1977), "Business cycle modeling without pretending to have too much a priori economic theory"

Procedure to compute a least squares regression with the covariance matrix proposed by Hodrick(1992) "Dividend Yields and Expected Stock Returns: Alternative Procedures for Inference and Measurement", Review of Financial Studies, vol 5, no 3, 357-386.

Example of user-defined menus.
RATS Version 8, User's Guide, Section 16.4

Analysis of linear regression with single break in time sequence.
This is based on an example from Stock & Watson, Introduction to Econometrics, 3rd edition, chapter 15.

Replication file for Ozbek and Ozlale(2005), "Employing the extended Kalman filter in measuring the output gap," Journal of Economic Dynamics and Control, vol. 29, no. 9, pages 1611-1622.
Demonstrates extended Kalman filter.

Generates the series of coefficients for an autoregressive representation from an input set of partial autocorrelations.

Procedure for Cointegration Tests in Heterogeneous Panels with Multiple Regressors ("Pedroni tests").

Pedroni (2004), "Panel Cointegration; Asymptotic and Finite Sample Properties of Pooled Time Series Tests with an Application to the PPP Hypothesis," Econometric Theory, vol. 20, 597-625.

Pedroni (1999) "Critical Values for Cointegration Tests in Heterogeneous Panels with Multiple Regressors," Oxford Bulletin of Economics and Statistics, vol. 61, 653-70.

Demonstrates various techniques for estimating linear models in panel data. Does fixed effects, random effects, first differenced and SUR using the instruction PANEL. Also shows how to implement create dummy variables and do panel data filtering techniques.

Example of panel causality test allowing for heterogeneity in the coefficients and variances.

Example of panel causality test allowing for heterogeneity in the coefficients and variances.

PANEDOLS.SRC is procedure for Peter Pedroni's methodology for group mean panel tests using Dynamic OLS.
Pedroni(2001) "Purchasing Power Parity Tests in Cointegrated Panels," Review of Economics and Statistics, vol. 83, 727-731.

PANELM.SRC is a new version of Peter Pedroni's methodology for group mean panel tests using Fully Modified OLS. Now in the form of the procedure, this is much easier to use than the old example program version (PANGROUP.PRG), and includes several important revisions by Prof. Pedroni.

Pedroni (2000) "Fully Modified OLS for Heterogeneous Cointegrated Panels," Advances in Econometrics, Vol. 15, 93-130.

Pedroni (2001) "Purchasing Power Parity Tests in Cointegrated Panels," Review of Economics and Statistics, vol. 83, 727-731.

This estimates Spatial Correlation Consistent (SCC) covariance matrix from panel data, from Driscoll and Kraay's "Consistent-Covariance Matrix Estimation With Spatially Dependent Panel Data", Review of Economics and Statistics article, November, 1998. It computes coefficient estimates and standard errors consistent for spatial correlation, autocorrelation and heteroskedasticity for linear instrumental-variable panel data. Also available are an example program using purchasing-power parity model (PPP01.PRG) and the accompanying data set (PPP.RAT).

Tests a fixed effects regression for one or two breaks in a single regressor, with breaks determined by the values of another variable.
Bruce Hansen(1999), "Threshold effects in non dynamic panels: estimation, testing and inference", Journal of Econometrics, vol 93, pp 345-368.

Demonstrates estimation of polynomial distributed lags using the instructions ENCODE and LINREG with UNRAVEL. See also PDLREG.SRC.

Automates the process of estimating Polynomial Distributed Lag models-- you list the variables, start and end lags, polynomial degree, select a constraint type, and the procedure does the rest.

Computes a Polynomial Distributed Lag. This is an update of the older PDL.SRC procedure included with RATS for many years. It offers more options, and the input conforms more closely to the standard form used by built-in RATS instructions.

Helps select a PDL model by computing AIC and BIC criteria. Note: this requires an older version of the PDL procedure (also originally by Norman Morin), which is available in the file PDLMORIN.SRC procedure.

Replication of Pedroni(2007), "Social capital, barriers to production and capital shares: implications for the importance of parameter heterogeneity from a nonstationary panel approach", Journal of Applied Econometrics, vol 22, no 2, 429-451. Uses the PANCOINT and PANELFM and procedures for testing and estimation of cointegration models in heterogeneous panels.

Replication file for Pedroni(2001) "Purchasing Power Parity Tests in Cointegrated Panels", Review of Economics and Statistics, 83, 727-731. Demonstrates the procedures panelfm.src and paneldols.src.

Performs the Perron test for a unit root allowing for a one-time change in the slope or level of the series. Useful when standard tests against trend stationary alternatives cannot reject the null hypothesis of unit root if the true data generating process is that of stationary around a trend function which contains a one time break. By Diego Vasquez, based in part on Norman Morin's URADF.SRC procedure.

Implements various unit root tests for series with endogenous time breaks, as described in Perron's paper "Further evidence on breaking trend functions in macroeconomic variables, Journal of Econometrics 80, 1997.

Implementation of the general structure for analyzing breaks with unit roots in Perron(2006), "Dealing with Structural Breaks," in Palgrave Handbook of Econometrics, Vol. 1, pp 278-352, extended to allow more than one break.

Computes one or more of the Perron-Ng "M" unit root tests.
Perron and Ng(1996), "Useful Modifications to Some Unit Root Tests with Dependent Errors and Their Local Asymptotic Properties", Review of Economic Studies, vol 63, 435-463.

Does the Perron-Rodriguez(2003) test for unit roots using GLS detrending, allowing for a break point at an unknown date.
Perron and Rodriguez(2003), "GLS Detrending, Efficient Unit Root Tests and Structural Change", Journal of Econometrics, vol 115, pp 1-27.

Replication file for Perron and Rodriguez(2003), "GLS Detrending, Efficient Unit Root Tests and Structural Change", Journal of Econometrics, vol. 115, no. 1, 1-27.

Replication file for Perron and Wada(2009), "Let’s take a break: Trends and cycles in US real GDP", Journal of Monetary Economics, vol 56, 749-765.

Uses frequency domain techniques to estimate the sum of the coefficients of the moving average representation for a series (persistence measure).

Replication file for Pesaran, Shin and Smith(1999), "Pooled Mean Group Estimation of Dynamic Heterogeneous Panels", JASA, vol. 94, no. 446, pp. 621-634. Demonstrates the instruction SWEEP.

Uses a multivariate Hannan's Efficient Estimator to estimate a set of linear equations with stationary errors and tests linear restrictions on the parameter matrix.

Phillips(1991), "Error Correction and Long Run Equilibria in Continuous Time." Econometrica, Vol. 59, 967-980.

Hall and Trevor(1992), "Long Run Equilibrium Estimation And Inference: A Non-Parametric Application." In P.C.B. Phillips (ed), Models, Methods and Applications of Econometrics: Essays in Honour of A.R. Bergstrom. Basil Blackwell.