Skip to content
Back to formatted view

Raw Message

Message-ID: <4FC7EBB6.6050406@xtra.co.nz>
Date: 2012-05-31T22:07:50Z
From: Rolf Turner
Subject: one more piece of info on AIC
In-Reply-To: <CAHz+bWYXn0fR9P+yXTb9SFi0NmMoahpejT8wb2=F9ViF=7B6Yw@mail.gmail.com>

On 01/06/12 02:52, Mark Leeds wrote:
> just one other thing about the AIC issue:
>
> there is a line in glm.fit which is the following:
>
> aic = aic(y, n, mu, weights, dev) + 2 * rank
>
> but I couldn't find the function aic so I couldn't investigate further. It
> looks suspicious though because it seems to me like
> it should be
>
> aic =  -2*likelihood + 2 * rank
>
> if anyone could help me find the aic function it's appreciated.

Dear Mark,

The aic() function depends on the family that you are using.

E.g.:

     binomial()$aic
     gaussian()$aic

     cheers,

         Rolf