Panel Data
Posted: Fri Aug 19, 2016 12:34 pm
Hi
I am running a panel regression of 16 funds over 16 years with 1/4 fund returns.
In the panel option I can choose fixed, random or non for both cross-sectional and Period.
I am confused as to which one to choose.
I understand that cross sectional fixed is variables that would vary across funds such management style ect..
and I understand that period fixed effects are variables that would vary over time such changes in the market.
I also understand that to choose between fixed and random you run a Hausman test which I have ran an I need to use "fixed"
But what happens or what results does eviews show me when I end up choosing cross section fixed or period fixed.
I am very confused on this matter.
Thank you.
I am running a panel regression of 16 funds over 16 years with 1/4 fund returns.
In the panel option I can choose fixed, random or non for both cross-sectional and Period.
I am confused as to which one to choose.
I understand that cross sectional fixed is variables that would vary across funds such management style ect..
and I understand that period fixed effects are variables that would vary over time such changes in the market.
I also understand that to choose between fixed and random you run a Hausman test which I have ran an I need to use "fixed"
But what happens or what results does eviews show me when I end up choosing cross section fixed or period fixed.
I am very confused on this matter.
Thank you.