12 Estimation of models
To use a model for forecasting, its parameters must be estimated from the data. Once a model has been identified (Section 11), two questions remain for an AR model, and the same two for any other:
- what is the order of the process;
- how can we estimate the parameters of the process.
The first was the subject of the previous section. This section deals with the second, starting with the simplest parameter of all, the mean.
12.1 Estimating the mean
12.2 Fitting an autoregressive model
12.3 Fitting a moving average process
12.4 Practice problems
12.2 Fitting an autoregressive model
12.3 Fitting a moving average process
12.4 Practice problems
Questions on this section
Stuck on something here? Ask below and it stays attached to this topic.