Question

Write two paragraphs about the forecasting method of Dynamic regression model. Discuss how to account for errors from a regression in an autocorrelation. What are those errors? How can they occur? Why do they matter? What can be done to deal with those errors to improve the quality of the forecasting model? Why don’t forecasters and managers look for or find these errors when using dynamic regression models?

Answer 1

In statistics, time series models can be divided into two groups:

1.Autoprojective time series models are models that involve only the time series to be forecasted (e.g. ARMA models).

Regression with ARIMA errors Regression models yt = β0 + β1x1,t + · · · + βkxk,t + et , yt modeled as function of k explanatory

2 Dynamic regression models are models that may involve the time series to be forecasted and the history of another time series as well.

There are basically two methods to reduce autocorrelation, of which the first one is most important:

1. Improve model fit. Try to capture structure in the data in the model. ...
2. If no more predictors can be added, include an AR1 model.
3. Causes of Autocorrelation
4. Inertia/Time to Adjust. This often occurs in Macro, time series data. ...
5. Prolonged Influences. This is again a Macro, time series issue dealing with economic shocks. ...
6. Data Smoothing/Manipulation. Using functions to smooth data will bring autocorrelation into the disturbance terms.
7. Misspecification.