Questions on Time Series data (i.year)

[Guest]

Dear all,

As far as I am concerned, there are 'norms' when we carry out regressions. For example, in my case at least, using -robust- to get rid of possible and potential heteroskedasticity is almost a 'default'. Do you think the command -i.year- should, in general, be included when we do regressions on time series data (e.g. stock market data) to detrend the data?

[Phil Bromiley]

You are unlikely to get a useful answer to such a general question. Some areas automatically put in year dummies, some don't. It would depend a lot on the situation.