No announcement yet.
  • Filter
  • Time
  • Show
Clear All
new posts

  • In and Out of Sample Forecast in SUR Model

    Hi, looking for some help on creating in sample and out of sample forecasts for my SUR model.
    I have data from 6 regions for the date range 1985h1 to 2017h2. Equation includes revenue as dependent variable, quantity and price as independent. All variables are logged, both independents also include first difference, lag, lead (see picture for 1 region's output). The inclusion of lead and lag is to allow model to be dynamic. (DSUR)
    I would like to create predicted values for revenue for each period for the entire frequency of the model and be able to display in a graph to illustrate model's effectiveness. Furthermore, I would like an in sample forecast fore the 5 most recent periods as well as an out of sample forecast for 5 periods.
    Stata will not let me use the "forecast" function with lead variables. I have begun using "predict" although the complexity of the model is causing me trouble. Any tips/advice?

  • #2
    Attached Files


    • #3
      You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Also, try to simplify your posting to the critical issues. Also, screen shots are discouraged.

      Have you looked at predict? If predict works and forecast doesn't, you might need to add observations with the right rhs variables to do future predictions. However, the idea of leads explaining something is odd.

      At the extreme, you can use the parameters to calculate your own predictions.