[R-sig-eco] Log transformed simple linear regression and Poisson regression

[Brooks,Mollie E]

Hi Lin,
If y is count data, then you should probably model it with either the 
Poisson or negative binomial distributions because these are for counts. 
You can sometimes get away with the normal distribution if the counts 
are very large.

There is an article by O'Hara and Kotz in Methods in Ecology and 
Evolution titled "Do not log-transform count data". So your first 
approach is probably unacceptable.

The similarity is that Poisson regressions typically use a log link 
function (the function glm does this by default). A log link means that 
the model follows the same exponential form that you've specified in 
your first approach (except for the +1). If y does not appear to have an 
accelerating relationship with the predictors, you can try an identity 
link, but it is more fragile.

cheers,
Mollie

On Thu, 2 Aug 2012 01:42:16 +0800 (CST), lgj200306 wrote:
> Hi, everyone,
> I used two methods to analysis the relationship between y (count 
> data) and x.
> 1) log transformed simple linear regression:
>>lm(log(y+1)~x1+x2+x3,data)
> 2) Poisson regression:
>>glm(y~x1+x2+x3,family=poisson())
>
> Some one told that these two ways are very similar, but other one
> told that the Poisson regression does not fit the y but the lambda
> (parameter of the poisson distribution). I am not sure which one is
> right. Can anybody help me?
> Thanks for your attention and best wishes for you!
>
> Lin
> Aug 1st, 2012
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology