[R] Power for Multiple regression?
s.nancy1 at yahoo.com
Wed Oct 7 16:12:33 CEST 2009
I came across this problem which is more a statistical question, I am hoping
that some one can clarify this for me
Is there a rule of thumb to determine how large your sample size should be
before you perform multiple regression. I have search the web and I found
some online tools where you provide the Alpha level, number of predictors,
the effect size and the desired power.
I have googled the effect size and I found that the effect size= R2/(1-R2)
Where R2 is the coefficient of determination.
I do not understand how to provide the effect size before doing the
regression and what it means? From my understanding, the power of a test is
the probability of rejecting a false null hypothesis. In this case what is
the hypothesis we are testing? Would the null be testing All the BETA
coefficients are ZERO simultaneously? Is it possible to look at the power to
test for each BETA separately? i.e H0: B=0 vr Ha: B not equal to 0
I appreciate your help
View this message in context: http://www.nabble.com/Power-for-Multiple-regression--tp25787420p25787420.html
Sent from the R help mailing list archive at Nabble.com.
More information about the R-help