Hi All,

I have been looking around the internet to see if I can undertake a panel data regression in excel but have not seen anything obvious. Everything I read says it can be done in STATA but I would still like to know if it can be done in excel. I want to do a fixed effects model:

yit = a + bxit + εit

Can I do this with Linest?

I have a panel of data for prices and demand for a good for a number of countries for a number of years and instead of modelling an individual elasticity for each country I would like to treat the data as a panel and get one elasticity.

Can the known y's in linest be two dimensional i.e. a seperate row for each individual countries demand tim series? Or do I have to put all the demand time series into one row somewhow? If the different country time series are in different rows as the known's can i still do multivariate regression i.e. a corresponding number of price rows for the different countries but some other explanatory variables too?

Thanks,

Eoin

## Bookmarks