GeneraliZd Linear Models: Goodness of fit

[Kornbrot, Diana]

HELP!

all sources I found give the following equations for converting -2log likelihood[-2LL] to other GoF measures that penalise for extra parameters

Bayes Information Criterion,  BIC = -2LL + K*ln(N)

Akaike InfromationCriterion, AIC = -2LL  + 2K

where K is number of extra independent parameters in the covariance matrix

N is the number of independent observations = Number of observations - number of estimated means

BIC given by SPSS is EXACTLY as defined above

How SPSS obtains AIC is a mystery. I was expecting AIC - (-2LL) to be an even integer, according to equation above

AIC - (-2LL) from SPSS is rarely an integer

Appreciate any help. Examples available on request

best

Diana

_____________________________________

Professor Diana Kornbrot
Mobile
[phone]
Work
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
[phone]
[hidden email]
http://dianakornbrot.wordpress.com/
http://go.herts.ac.uk/Diana_Kornbrot
skype:  kornbrotme
Home
19 Elmhurst Avenue
London N2 0LT, UK
[phone]                                                   
 ------------------------------------------------------------                                    

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Jon Peck]

To within rounding error, I see your expression as an even integer if I use the exact values in the output (double click table and click in cell to see the full precision values).  E.g., 6.000001 for one example.

On Wed, Jul 25, 2018 at 3:38 AM Kornbrot, Diana <[hidden email]> wrote:

HELP!

all sources I found give the following equations for converting -2log likelihood[-2LL] to other GoF measures that penalise for extra parameters

Bayes Information Criterion,  BIC = -2LL + K*ln(N)

Akaike InfromationCriterion, AIC = -2LL  + 2K

where K is number of extra independent parameters in the covariance matrix

N is the number of independent observations = Number of observations - number of estimated means

BIC given by SPSS is EXACTLY as defined above

How SPSS obtains AIC is a mystery. I was expecting AIC - (-2LL) to be an even integer, according to equation above

AIC - (-2LL) from SPSS is rarely an integer

Appreciate any help. Examples available on request

best

Diana

_____________________________________

Professor Diana Kornbrot
Mobile
[phone]
Work
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
[phone]
[hidden email]
http://dianakornbrot.wordpress.com/
http://go.herts.ac.uk/Diana_Kornbrot
skype:  kornbrotme
Home
19 Elmhurst Avenue
London N2 0LT, UK
[phone]                                                   
 ------------------------------------------------------------                                    

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

--
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD