Dear everyone,
I am currently working on my first timeseries homework. The main task is to estimate an AR(1)-GARCH(1,1) model for a small series of price returns. I did this with the following type of code:
arch log_returns, arch(1) garch(1)
The estimated results seem plausible. Now my problem. The next task tells me to use the results to predict one day ahead values for the price returns as well as the variance.
When I use the predict command, it gives me the same value for all the ahead prices (which does not seem very plausible to me), while giving back likely predictions for the volatility.
My main question is now, how can I generate the right predicted values? Do I have to use the forecast command and if so, how does it work for the AR(1)-GARCH(1,1) model?
I would be delighted if anyone could help me with this issue.
Many thanks in advance.
I am currently working on my first timeseries homework. The main task is to estimate an AR(1)-GARCH(1,1) model for a small series of price returns. I did this with the following type of code:
arch log_returns, arch(1) garch(1)
The estimated results seem plausible. Now my problem. The next task tells me to use the results to predict one day ahead values for the price returns as well as the variance.
When I use the predict command, it gives me the same value for all the ahead prices (which does not seem very plausible to me), while giving back likely predictions for the volatility.
My main question is now, how can I generate the right predicted values? Do I have to use the forecast command and if so, how does it work for the AR(1)-GARCH(1,1) model?
I would be delighted if anyone could help me with this issue.
Many thanks in advance.
