I´m interested to know if it´s possible (and what it means) to obtain a significant linear model regression (p value=0.008) with coefficients fitted are not significant (p>0.05)?
Asked
Active
Viewed 44 times
0
-
You will find extensive discussions of this and related phenomena by searching related keywords that include collinearity. – whuber May 03 '13 at 20:42
-
That's interesting @Dimitriy, but I feel obliged to point out it's the reverse situation: the regression itself is not significant even though individual coefficients are "significant." – whuber May 03 '13 at 20:45