The attached zip has the examples and data files from West and Harrison,
Bayesian Forecasting and Dynamic Models, 2nd ed, Springer 1997. These are all examples of the use of the
DLM instruction for analyzing state space models. The authors take a very different approach however, as they emphasize the use of state space models for forecasting with small data sets with possibly changing conditions. Unlike the Durbin and Koopman, Commandeur and Koopman, and Harvey books, they use informative rather than diffuse pre-sample information, and make use of "discounting" for handling the variance of the increment in the state equation which multiplies up the existing uncertainty rather than using the standard additive change to the variance.
| Example | Description | RATS Level |
| westp040.rpf | Local level model with variance change | Intermediate |
| westp057.rpf | Local level model diagnostics, forecasting | Advanced |
| westp081.rpf | Time-varying parameters regression | Intermediate |
| westp082.rpf | Time-varying parameters regression | Intermediate |
| westp084.rpf | Time-varying parameters regression | Advanced |
| westp257.rpf | UC model with seasonals | Advanced |
| westp318.rpf | UC model with seasonals | Advanced |
| westp387.rpf | UC model with seasonals | Advanced |
| westp434.rpf | State space model, Bayesian analysis | Advanced |
west-harrison.zip