Expected return model, do we delete days with zero trading volume?
Posted: 23 June 2014 08:10 PM   [ Ignore ]
Jr. Member
Total Posts:  34
Joined  2013-03-31

Hi Joost,

I am trying to calculate abnormal returns for an event study.

To get abnormal returns, I first use historical data to estimate an expected return model: y=a+beta1*b,

When estimating the expected return model, do we only keep observations with non-zero trading volume?

Thanks a lot,


Posted: 24 June 2014 07:59 AM   [ Ignore ]   [ # 1 ]
Total Posts:  901
Joined  2011-09-19

hi Anna,

In the grand scheme of things, it probably doesn’t matter.Results for raw returns, size adjusted returns, market model, 3-4 factor models; they are typically consistent with each other (the measures are highly correlated).

When there is no trade on a given day, the (end of day) stock price and return are based on the mid-point of the bid and ask.  So non-trading days (months) still have return measure. One way to mitigate the issue is to use monthly returns to estimate the expected return model.

Hope this helps,




To reply/post new questions: Please use the group WRDS/SAS on Google Groups! http://groups.google.com/d/forum/wrdssas