Page 1 of 1
regression with population
Posted: Wed May 19, 2010 11:57 pm
by joseph
Data is a population, need test assumptions (E(u)=0, u is normality distribution,...) OLS regression with population?
regression with population
Posted: Thu May 20, 2010 6:48 am
by EViews Gareth
I'm not sure I understand what your question is.
Re: regression with population
Posted: Thu May 20, 2010 8:30 am
by joseph
I imply that my data set is observed on population. Is test assumptions of residual OLS required for PRF-Population Regression Function?
Re: regression with population
Posted: Thu May 20, 2010 10:20 am
by startz
I imply that my data set is observed on population. Is test assumptions of residual OLS required for PRF-Population Regression Function?
The usual assumption for the statistical properties of OLS are that the data is draws from an infinite population. If you have the entire population, what is the statistical question?
By the way, in general it is not possible to test E(u)=0 and the assumption that u is normal is not important for most regression results.