Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Poisson

By N | Apr 26, 2016 04:00AM CEST

Which R2 should i use for model evaluation?
Goodness of fit statistics (Variable Mortality (thous. heads)):

Statistic Independent Full
Observations 15 15
Sum of weights 15.000 15.000
DF 14 8
-2 Log(Likelihood) 191.387 152.940
R²(McFadden) 0.000 0.201
R²(Cox and Snell) 0.000 0.923
R²(Nagelkerke) 0.000 0.923
AIC 193.387 166.940
SBC 194.095 171.897
Deviance 41.519 3.073
Pearson Chi-square 56.158 3.089
Iterations 0 6

Up

0

Down

By Efthalia | Apr 26, 2016 10:39AM CEST | XLSTAT Agent

Hello,

You can use any of these pseudo R2 to measure how well the model is adjusted. R² (McFadden) is commonly used. However opinions vary in the statistical community about which one is considered as the best one.

Another important value to look at is the probability of Chi-square test on the log ratio. This is equivalent to the Fisher’s F test: we try to evaluate if the variables bring significant information by comparing the model as it is defined with a simpler model with only one constant.

Regards,
Thalia

This question has received the maximum number of answers.

Contact Us

1c26995d494fb3061dd0ae8571ffc0a4@xlstat.desk-mail.com
https://cdn.desk.com/
false
desk
Loading
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
about
false
Invalid characters found
/customer/portal/articles/autocomplete
9283