Ali,
The size of the data set is not an issue. If your residuals are not normally distributed (or close to normally distributed) you risk making incorrect inferences.
John
John Sorkin M.D., Ph.D.
Chief, Biostatistics and Informatics
Baltimore VA Medical Center GRECC,
University of Maryland School of Medicine Claude D. Pepper OAIC,
University of Maryland Clinical Nutrition Research Unit, and
Baltimore VA Center Stroke of Excellence
University of Maryland School of Medicine
Division of Gerontology
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)
[log in to unmask]
>>> Ali Khashan <[log in to unmask]> 4/2/2007 8:24 AM >>>
Dear all,
I am using a very large dataset (N=1.3m) to analyse
the effect of a categorical variable on gestational
age at delivery (continuous variable). I used linear
regression
(regress in stata) but as you would probably expect
the residuals are not normally distributed. I used
qreg in stata but the results do not look right
(probably there is a convergence problem).
Few months ago someone suggested that if we have a
large dataset we can use linear regression even if the
residuals are not normally distributed. Is this right?
If yes is there any reference?
Also, I could not find a meaningful transformation for
gestational age and cox regression would not work
because I need the effect of exposure on continuous
gestational age.
Any help is highly appreciated
Best Wishes
Ali
___________________________________________________________
The all-new Yahoo! Mail goes wherever you go - free your email address from your Internet provider. http://uk.docs.yahoo.com/nowyoucan.html
Confidentiality Statement:
This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.
|