I want to create variable_3 taking the difference of variable_1 and variable_2
in order to run the OLS regression containing variable_3, correctly, what should I do for the dates that variable_1/variable_2 don't have data?
If there are a few missing dates, you can ignore them.
Another simple alternative is using the average of neighboring points as a candidate instead of the missing value.
Honestly this looks more like a repeated measures linear model to me rather than a true time series. I would suggest you investigate an ANCOVA approach where your model would be v3= a + b*x1 +c*x2+ ....if you google ANCOVA regression analysis you should find lots of info on the ols approach to your regression add missing data to your query and that will give you options depending on your fav stat package. Best wishes, David Booth
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series