Hi,
Can anyone please point me to any websites (or books) which they would
recommend that take you through the diagnostic checks (via examples) one
should do on statistical models (anything on Logistic, Poisson, Negative
Binomial or Cox Proportional Hazard would be great).
As well as the basics, I'm interested in questions such as if outlier
residuals are found (with a more extreme value that you would expect given
the underlying distribution and sample size) and if after checking you find
the data behind these residuals is correct what should you do - take these
observations out of the analysis (and describe their characteristics in an
accompanying report) and re-estimate or should you just leave them in. Even
if these outlying residauls have low leverage and thus their omission does
not effect the model estimates greatly I'd still like to know if there is
some sort of consensus as to what one should do.
Any help much appreciated - I'll post the main responses.
Regards,
Steve
DISCLAIMER: The information in this message is confidential and may be
legally privileged. It is intended solely for the addressee. Access to this
message by anyone else is unauthorised. If you are not the intended
recipient, any disclosure, copying, or distribution of the message, or any
action or omission taken by you in reliance on it, is prohibited and may be
unlawful. Please immediately contact the sender if you have received this
message in error. Thank you.
|