I have monthly time series on forecasts (for the months of August, September, October, November, December, and January.) The data is only available for these months and doesn't exist for other months. I am trying to see how a shock affects the forecasts using impulse response functions. However, I was asked a question during an academic presentation that I can't use IRFs since they are discontinuous time series. However, I am unable to find anything on this in the books/ journal articles. I am inclined to argue that we can still use IRFs for my data. Can anyone tell if this is correct? If not, how do I approach this? Is there any research that uses IRFs on discontinuous time series?
$\begingroup$
$\endgroup$
Add a comment
|