Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

Logistic Regression results, Please Help. Thank You

Reply
Frequent Contributor
Posts: 95

Logistic Regression results, Please Help. Thank You


Hi All,

I have  built a Logistic Regression Model and I get the results below...The variables with a high Chi-Square below (Var1 and Var6)..Should they be removed from the model?

Thank You

Analysisof MaximumLikelihoodEstimates
ParameterDFEstimateStandard ErrorWald Chi-SquarePrPr> ChiSqStandardized EstimateExp(Est)
Intercept1-2.81450.004495,784.70 <.0001 0.06
Var 110.07580.00035745,168.02 <.00010.26611.079
Var 210.36460.007532,345.72 <.00010.04031.44
Var 31-0.09120.001862,407.66 <.0001-0.0520.913
Var 410.78910.008099,506.31 <.00010.09812.201
Var 510.10890.003341,060.29 <.00010.03391.115
Var 611.0980.00339104,610.00 <.00010.70952.998
Var 710.1530.002394,092.80 <.00010.0621.165
Super Contributor
Posts: 336

Re: Logistic Regression results, Please Help. Thank You

I think a statistically right way would be, if you run the regression again - without var1 & var6 - and compare the likelihood functions. If they don't differ much, you don't need the 2 variables. (Likelihood-Ratio-Test: Subtract lambda=2*(likelihood with var1&var6 - likelihood without var1&var6) -> get value of the chi-squared distribution with 2 degrees of freedom -> if the result is almost zero (e.g. <0.01) the 2 variables should not be excluded.)

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