Forecasting with preliminary data: a comparison of two methods

This study examines two alternate methods, a vector autoregression error co rrection model and a state space model, to forecast revised United States t rade balance figures. Both these methods incorporate preliminary and revise d trade data. The results obtained from these methods were compared to the benchmark forecasts generated by revised-data-only models. This Study finds that the state space model performs worse than the benchmark. The vector a utoregression model performs better than the benchmark only in the one-step forecast. These results indicate that incorporating preliminary data may n ot be useful in forecasting the revised data.