Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Coefficient estimates of GLM
Message posted by Y Ye on February 16, 2000 at 12:00 AM (ET)
Hi everyone,
I am doing some Generalized Linear Modelling.However, I do know how to view the coefficient
estimated derived from GLM. For example, I have a factor "Year" which has 10 years. I get 10 coefficient
estimates, i.e. Intercept, year1, year2,...year9. I explained that the "intercept" is the mean of the first
year, "year1" is the mean of Year 2 relative to Year 1. The absolute mean of Year 2 is "Intercept + year1" and so on.
Am I right? However, the softwar, S-Plus, also provide a graph showing the year effects. I found that the coefficient
estimates and the graph of effect show different patterns. I do not know which one I should take? How should be explain
all the outputs of the software.
Thank you very much in advance for your time and help.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.