Jennifer Devine wrote:
I'm running a GLM on fish catches that were collected with different gears
over time. It was suggested to me that if area swept differed between the
gears, I would need to adjust for that by using an offset with the offset =
area swept by the gear. However, I'm running into problems on how to set
this up. I've made a new variable in my data called offset and for each
gear type, I have replace offset=area swept for that gear. Area swept is in
sq. nautical miles and is less than one. My model is: glm catch yr strat,
f(nb) link(log) off(offset). This is where I run into problems - is my
offset correct? My data, visual inspection, linear regression, GLM, all
show a decline over time without the offset. Once the offset is added, my
results are a mess. Can someone please help?
--------------------------------------------------------------------------------
Try either taking the logarithm of the area swept before including it
as -offset()-, or including the untransformed area swept as -exposure()-
instead of -offset()-.
Joseph Coveney
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
