This page provides a bug-fix history for Version 4 of RATS. For bugs and bug-fix information for Version 5, please see the Known Bugs page.

Software bugs are a fact of life, although we make great efforts to minimize them. By making a list of known bugs available, we hope to minimize the inconvenience these problems may cause. With that in mind, we present the much belated "list of known bugs in RATS".

For software like RATS, these can be grouped in three basic categories:

• Serious: Bugs that produce incorrect results that are not obvious to the user, and/or that impose major limitations on functionality.

• Annoying: Bugs that are obvious but annoying, or impose moderate limitations.

• Minor: Minor or very rare inconveniences, presenting little or no limitations on functionality (i.e. interface glitches, problems with easy workarounds, etc.).

This is the "long version" of our known-bugs list. It includes all known the "serious" and "annoying" bugs, and most of the minor ones. We have omitted a few categories of minor bugs (interface problems, some rare system-specific issues, etc.) to make the list as useful as possible to the widest audience.

For a shorter version of this list that includes only the "serious" bugs, see Known Bugs (Short Version).

For the most part, this only includes bugs that were present in release 4.10c or later (4.10c was a bug-fix update to Version 4.10). If you are using something older than that, we would recommend updating to get the bug fixes mentioned below (in addition to all the spiffy new features, of course). At some point we will extend this list to include older problems, particularly those affecting Mac users stuck at Version 4.02. Stay tuned.

Note: As of May, 1998, the current version of RATS for all platforms except Mac is 4.31, which is identical to 4.30 except for the bug fixes noted below.

Bugs Not Fixed as of 4.31

• If you use the COVMAT option on RESTRICT (with CREATE), the covariance matrix is set properly, but the standard errors displayed in the output are based on the new %SEESQ and the old (unrestricted) %XX matrix. Workaround: omit COVMAT option, use COMPUTE to set a variable equal to (the new) %XX after the RESTRICT.

• Due to a bug in the compiler library, the %DENSITY or EXP functions will return an incorrect answer for very small arguments. For example, the answers returned by %DENSITY(1.05e-15) and %DENSITY(0) will be different. Workaround: Do something like:

  COMPUTE EPS = 1.0E-11
  COMPUTE D = %IF( ABS(X)<=EPS, %DENSITY(0.0), %DENSITY(X))
  

• Only under Windows NT, WinRATSDATA doesn't automatically make new windows "active", which can lead to crashes if you try to do Import with no "active" window. Workaround: just click anywhere on the new window to make it active before working with it.

• STWISE with METHOD=BACKWARDS and FORCE=n options can occasionally cause a crash (output prior to crash shows multiple copies of the p value for one parameter, with the p-value equal to 1.0).

• Adding FIND to TVARYING to search for hyperparameter values can cause garbage output or crash.

• Problem reading data from Portable format files if you specify a range shorter than that available on the file (generates bogus IO30 error message, skips next series). Workaround: read in all the data, convert it to RATS format.

• If you use RATSDATA to graph two or more series which begin after the first subperiod of the year (i.e. in the 2nd month or later, 2nd quarter or later, etc.), one of the series is shifted several observations to the right.

• High-Low-Close graph on WinRATSDATA does not function properly. Use DOS version (or RATS).

• Contrary to the documentation, the %RIGHT, %LEFT, and %MID functions currently only support LABEL variables (STRINGS are truncated to 16 characters).

• Using STATS with the FRACTILES option crashes WinRATS-32 if the series is very large.

• Using "*" to skip path series when using the PATHS option on FORECAST fails with an SC4 error message when used in compiled sections (e.g., inside a loop). Workaround is to list the equations in such an order that you don't need to skip any (i.e. so that the equations that need paths are listed first).

• The DATA instruction can't read back in panel data series written with COPY(FOR=PORTABLE).

• Doing COMPUTE A = B where B is a VECTOR of arrays (e.g., VECT[RECT]), this makes A simply point to the same memory space as B, which can cause unexpected results. Better is to dimension A and use EWISE instructions to do elementwise copying of values from elements of B to elements of A.

• INPUT fails in compiled sections (procedures, loops, etc.). Use READ or COMPUTE instead.

• The program currently isn't designed to allow you to use LOCAL variables as NONLIN parameters. If you try, the program will usually crash. This feature may be supported in Version 5.

• With GRAPH(STYLE=STACKEDBAR) and 2 series, RATS uses the same color or pattern for both series for entries where one series is positive and one series is negative.

Bugs Fixed in Version 4.31

• Equality constraints frequently would not hold in LQPROG problems (RATS would compute a solution treating the equality constraints as <= constraints, rather than equality constraints). Appeared in 4.30. Fixed in 4.31.

• In NNLEARN, changed the way node values were scaled to eliminate some convergence problems. Appeared in 4.30. Fixed in 4.31.

• The formula we used for KAPPA with the cosine taper was wrong—Pi should not appear in either the numerator or the denominator. Present since 4.0. Fixed in 4.31.

• Long expressions in SET, COMPUTE, or FRML instructions could cause bad behavior in RATS 4.3 (crashes, hangs). Wouldn't usually appear in the actual instruction with the long expression, but would eventually hang. Fixed in 4.31.

• OVERLAY instruction would fail with syntax error on Sun Solaris systems. Fixed in 4.31.

• With UNIX RATS, if you used the PATTERNS option on a GRAPH with N series, only N-1 series would show up on the screen. All N series would be saved to the graph in the RGF file, however.

• Doing %INVNORMAL(X) where X was very small (<1.0e-10) produced an NA where it should have produced a valid result. Fixed in 4.31.

• When printing to LaserJet printers using PCL (i.e. non-PostScript) drivers, WinRATS may skip one line of output at the bottom of each page. Not a problem when printing using PostScript drivers. Workarounds: save and import or copy and paste into a word-processor or similar program and print from there. Fixed in 4.31 (Note added June 26, 1998).

Bugs Fixed in Version 4.21 or 4.30 (4.30 for UNIX)

Versions 4.21 (PC platforms only) and 4.30 were released at approximately the same time. 4.21 added only bug fixes and was available for shipping/handling costs, while 4.30 include the same bug fixes plus new functionality and was slightly more expensive ($25 from 4.2 to 4.3).

• Original release of WinRATS-32 4.20 would not start up normally under Win95 or NT 4.0 (which shipped after RATS 4.20). Workaround was to run as minimized icon (i.e. start in batch mode, then switch back to interactive using File-Batch Mode). Fixed in August, 1995 build of 4.20.

• RATS would crash if you tried to copy a graph that included VLABELs or OVLABELs. Fixed in Version 4.21.

• Problems copying and pasting graphs under Windows NT. Fixed in April, 1997 build of 4.21.

• Due to some algorithm changes, the NOSQUARE option on MCOV has no effect in Version 4.21 (i.e. produced same results as if you had used default SQUARE setting). This affects, for example, our ROBUST.PRG example (page 5-11). As long as the resids series is all positive, you can work around this by just taking the square root of the series before applying MCOV. Fixed in 4.30.

• If you use AR1 with panel data, the residuals are not set correctly. Rather than being "NA", the first obs. of each indiv. will be set to the previous residual value. Fixed in 4.21.

• %MQFORM(A,B) produced wrong answers when the A matrix was declared as RECTANGULAR. Worked properly when A was SYMMETRIC (workaround is to just use the standard matrix functions for which %MQFORM is a shortcut—see Appendix B). Fixed in 4.21.

• The DAMP option had no effect on NLSYSTEM (so standard errors were, if you will, the "undamped" standard errors). Fixed in 4.21.

• The TEST instruction returned a Chi-squared statistic when used after a BOXJENK—should have returned an F-statistic. Fixed in 4.21.

• Through 4.20, MVFRACTILE used the wrong weights in most cases when interpolating between two numbers (i.e. if the fractile doesn't hit an exact data point). Numbers were close (between the correct two numbers, but weighted incorrectly). MEDIAN option worked OK, but most other factors would give the wrong numbers. Fixed in 4.21.

• You couldn't save changes to existing RATS format data files (this appeared only in the August, 1995 build of WinRATS-32 4.20). Fixed in 4.21.

• EXCLUDE and SUMMARIZE would not work after an AR1 instruction. Fixed in 4.21

• MVFRACTILE frequently crashed RATS if the new series parameter wasn't used. Fixed in 4.21.

• Problem with ENABLE and CHECK options on USERMENU when using more than two USERMENUs. Fixed in 4.21.

• Year labels were sometimes wrong when graphing very long monthly data series. Fixed in 4.21.

• When doing bar graphs with series that contained only negative values, bars were filled going up from the bottom of the graph, rather than down from the 0-axis. Appeared in 4.20. Fixed in 4.21.

• If you use a FORTRAN format with 5 or less columns on a COPY instruction (possibly other instructions as well), missing values are printed as "NN" instead of "NA." With 6 or more columns, the "NA" is displayed correctly. Appeared in 4.20. Fixed in 4.21.

• Couldn't read data from CITIBASE/DRI Basic Econ database by assigning the appropriate LABEL to a numbered series. It will work if you assign a name with EQV, however. This was due to a upper vs. lower case problem. Fixed in 4.21.

• The NAMES='...' option on KFSET and TVARYING did not work properly (returning an SX22 error message). No known workaround, other than to use a the list of covariance matrices parameter instead. Fixed in 4.21.

• The MVFRACTILE instruction (at least with the MEDIANS option) crashes all versions of RATS 4.20 unless you supply a name for the "newseries" parameter. Fixed in 4.21.

• The APPEND option on OPEN did not work properly in 4.20--it simply replaced the existing file with a new file. Fixed in 4.21.

• The KLABEL option didn't work properly on SCATTER graphs in 4.20. Fixed by the August 10, 1995 build of WinRATS-32 (the Win95 fix release).

• By design, RATS will automatically switch to using DOTS on a SCATTER plot above a certain number of observations, even if the instruction specifies STYLE=SYMBOL. However, in WinRATS 4.20, the dots were not displayed on the screen (did print properly). Fixed in 4.21.

• When running WinRATS in batch mode, the input lines of the first program executed run were not echoed to the output file, until an instruction that actually produces output was executed. Fixed in 4.21.

• Some problems with header placement on WinRATS when doing GRAPHS after GRPARM(PORTRAIT). Fixed in August, 1995 build.
• A perfect-fit regression, such as a constant on CONSTANT will crash WinRATS-16 if Q-stats are calculated (i.e. if a CAL is in effect). WinRATS--32 returns NA's for Q-stat and significance level. Fixed in 4.21.

• Problems with handling the HEADER option on COPY (in 4.0-4.1, didn't wait for header line in interactive mode, in 4.2/4.21 ignored header line completely). Fixed in 4.30.

• After encountering a syntax error because of missing close quote (') on a string, RATS would not accept the $ (line-continuation symbol) symbol. Would have to quit and re-start to eliminate the problem. Fixed in 4.20.

• When saving undated panel data to a RATS format file, there was a problem with handling missing values at the beginning of the data set (the NA's weren't saved, so the data was shifted "up"). Fixed in 4.20.

• The FRML option on an EQUATION instruction would associate a FRML properly, EXCEPT that the dependent variable was not defined, which could generate NL5 error on forecasting. Fixed in 4.20.

• Through 4.20, ESTIMATE would crash the program if the SYSTEM was defined using the CMOMENT option. Workaround: Just omit the CMOM option. Fixed in 4.21.

Bugs Fixed in Version 4.20

• Standard errors reported by MAXIMIZE were incorrect when using the ROBUSTERRORS option. Note that BHHH results should be unaffected by ROBUST, but, due to this bug, ROBUST produced different std. errors than NOROBUST. Std. errors also wrong for BFGS with ROBUST. Fixed in 4.20.

• Leading NA's and trailing NA's in a panel data set get chopped off when doing the PANEL instruction if you use a / for default range (i.e. the data would be shifted "up", or earlier in time). Workaround is to use explicit date ranges for start and end. Fixed in 4.20.

• Through 4.10, EIGEN instruction did not sort results correctly if the input matrix was type RECTANGULAR (non-symmetric) matrix. Problem fixed in 4.20.

• For univariate models, %LOGDENSITY would return the determinant, not the LOG(DET), of the first term, and would this give incorrect results. Works OK for multivariate, not univariate. Workaround is to write them the old fashioned way. Fixed in 4.20.

• Through version 4.10, ORDER(ALL) didn't sort the last (highest numbered) series. All series are sorted if you list them explicitly (e.g. "ORDER X / Y Z" sorts X, Y, and Z, while "ORDER(ALL) X" sorts only X and Y). Fixed in version 4.20.

• Through version 4.10, STWISE with the METHOD=BAKCWARD option would always use the P-values from the initial (full) regression, except for slight differences due to changing degrees of freedom. Thus STWISE could very well report the wrong final model. Fixed with version 4.20.

• Problem when using CMOM and LINREG(CMOM) inside a DOFOR loop if the loop index was defined as a series (obviously weird regression results). Worked fine if the DOFOR loop was declared as an integer, rather than a series. Fixed in 4.20.

• RATS would crash on a GRAPH instruction with VTICKS=N if N was 3 or less (i.e. VTICKS=4 or greater works OK). Fixed in 4.20.

• If you use ENVIRONMENT RATSDATA=filename, RATS wouldn't let you read data from the file (you could edit, store, save, etc. as expected). Fixed in 4.20.

• Problem with converting data from RATS format to RATS format using STORE(CONVERT=RATS) (some data would be slightly incorrect due to bogus frequency conversion attempt). Fixed in 4.20.

• When doing SCATTER with (STYLE=LINE) and 2 or more pairs of series, the key would properly show different line patterns, but lines in the graph would be drawn using the same pattern (solid line). Fixed in 4.20.

• When doing COPY(FOR=WKS,DATES), RATS 4.10c sometimes writes out the dates incorrectly for monthly, quarterly, and annual data (daily, sevenday, weekly are all OK). Fixed in 4.20.

• RATS ignored * symbols on "PATHS" supplementary cards for impulse, so you may get shocks applied to the wrong equation if skipping some equations. Workaround is to use 0 (the number zero) instead of * to skip equations. Fixed in 4.20.

• When using GRAPH(GRID=...), the grid lines are drawn at the tick marks, rather than through the data point (which usually lies between tick marks). Fixed in 4.20.

• If an individual (or time period) in a panel set is all NA (missing values), PANEL(COMPRESS) sets the corresponding entry in the output series to 0, rather than NA. Fixed in 4.20.

• SCATTER plot symbols were not converted when exporting to PIC or HPGL. Thus PostScript was the only way to import scatter plots into word-processors, etc. Fixed in 4.20.

• TYPE option choices on TAPER were not recognized. Fixed in 4.20.

Bugs in Older Versions

Most bugs fixed by Version 4.10c or earlier are not listed on this page. However, a few important ones are shown below:

• Through version 4.02, hypothesis tests done after an NLSYSTEM or MAXIMIZE estimation were done as F-tests, when they should have been chi-square tests. Fixed with Version 4.10.

• When doing EXCLUDE after doing AR1, the statistic used for the test is wrong by a factor of %SEESQ. Workaround is to take chi-squared statistic from output and divide this by %SEESQ to get the proper statistic, than apply CDF:

  COMPUTE STAT = # from Exclude output/%SEESQ
  CDF CHISQ STAT 1
  

  Fixed with 4.10

• Weighted Least Squares Problem:

  INST Z1 Z2
  LINREG(INST,SPREAD=S) Y
  # X1 X2 
  

  yields the wrong results. Instead use

  INST Z1 Z2
  CMOM(SPREAD=S)
  # Z1 Z2
  COMPUTE %CMOM = INV(%CMOM)
  LINREG(INST,SPREAD=S,WMATRIX=%CMOM) Y
  # X1 X2
  

  Fixed in RATS 4.10

---

Copyright © 2013 Estima. All Rights Reserved.
This file was last modified on 05/13/09