[R] statistical question: confidence interval of regression weight - significance

Martin Batholdy batholdy at googlemail.com
Sun Mar 8 03:01:51 CET 2009


at first; thanks for the help on getting confidence intervals in R.

now I have a pure statistical question.
I hope you don't mind if I ask ...

I have an expectation of how large my beta-weight in a regression  
should be - so I have an "ideal" or expected regression line.

Now the real beta-weight is less then the expected and when I draw the  
confidence interval lines above and below the estimated line, the  
expected regression line is outside of the confidence intervals at the  
near end and the beginning of the x-scale.

Can I say know that the empirical beta-weight is significantly  
different from the expected beta weight?

Or is there a test (perhaps implemented in R) that can tell how the  
probability is that the estimated (empirical) beta-weight comes from a  
population where the beta-weight is equal the expected?

thanks for any help!

More information about the R-help mailing list