Proc Logistic - Hosmer and Lemeshow Goodness-of-Fit

Reply
Occasional Contributor
Posts: 14

Proc Logistic - Hosmer and Lemeshow Goodness-of-Fit

Hi,

anybudy can guide me if my Hosmer and Lemeshow Goodness-of-Fit output is as bellow.... how it will have impact on model?

Hosmer and Lemeshow Goodness-of-Fit
Test
Chi-SquareDFPr > ChiSq
363.68388<.0001

Othe SAS output is good as below only Hosmer and Lemeshow Goodness-of-Fit is not good:

Association of Predicted Probabilities and Observed Responses
Percent Concordant85.2Somers' D0.705
Percent Discordant14.7Gamma0.706
Percent Tied0.1Tau-a0.33
Pairs1.37E+09c0.853

Testing Global Null Hypothesis: BETA=0
TestChi-SquareDFPr > ChiSq
Likelihood Ratio29145.5312<.0001
Score14436.3512<.0001
Wald18640.1812<.0001

Fit Statistics for SCORE Data
Data SetTotal FrequencyLog LikelihoodError RateAICAICCBICSCR-SquareMax-Rescaled R-SquareAUCBrier Score
WORK.SAMPLE76553-360630.2272152.0172152.0172272.272272.20.3166330.4315940.8525730.150807
WORK.HOLD_OUT32802-155940.221731213.9531213.9631323.1331323.130.310730.4235440.8524830.151017

Thanks,

KP

Contributor jf
Contributor
Posts: 22

Re: Proc Logistic - Hosmer and Lemeshow Goodness-of-Fit

HL test in proc logistic uses 10 groups to measure the goodness of fit, i.e, the kai square test always has 8 degrees of freedom. However, based on the predictors, the number of  natural groups is usually not 10, i.e. the test statistics doesn't follow the distribution of kai square (8) .

I don't look at the statistics usually. Your C value is 0.85, that is pretty good. I think you got a good model.

Ask a Question
Discussion stats
  • 1 reply
  • 339 views
  • 0 likes
  • 2 in conversation