In: Computer Science
Review the predictor variables and guess from their definition at what their role might be in a credit decision. Are there any surprises in the data?
##
## Call:
## glm(formula = response ~ chk_acct + duration + credit_his +
purpose +
## amount + saving_acct + installment_rate + sex + other_debtor
+
## age + other_install + telephone + foreign, family =
binomial,
## data = german_credit.train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.2263 -0.7082 -0.3799 0.6923 2.7670
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.608e+00 8.251e-01 1.949 0.051344 .
## chk_acctA12 -3.628e-01 2.408e-01 -1.507 0.131894
## chk_acctA13 -9.224e-01 3.882e-01 -2.376 0.017499 *
## chk_acctA14 -1.942e+00 2.623e-01 -7.404 1.32e-13 *
## duration 2.516e-02 1.000e-02 2.516 0.011868 *
## credit_hisA31 -6.266e-01 6.195e-01 -1.012 0.311749
## credit_hisA32 -1.190e+00 4.761e-01 -2.500 0.012422 *
## credit_hisA33 -1.118e+00 5.413e-01 -2.065 0.038939 *
## credit_hisA34 -1.695e+00 5.007e-01 -3.386 0.000710 *
## purposeA41 -1.437e+00 4.026e-01 -3.568 0.000359 *
## purposeA410 -2.149e+00 9.172e-01 -2.343 0.019151 *
## purposeA42 -8.366e-01 2.850e-01 -2.936 0.003327 **
## purposeA43 -9.470e-01 2.748e-01 -3.447 0.000567 *
## purposeA44 -5.495e-01 7.469e-01 -0.736 0.461921
## purposeA45 -2.052e-01 6.364e-01 -0.322 0.747138
## purposeA46 2.944e-01 4.299e-01 0.685 0.493510
## purposeA48 -2.177e+00 1.216e+00 -1.790 0.073463 .
## purposeA49 -8.948e-01 3.800e-01 -2.355 0.018531 *
## amount 1.549e-04 4.937e-05 3.137 0.001704 **
## saving_acctA62 -5.153e-01 3.222e-01 -1.599 0.109744
## saving_acctA63 1.053e-01 4.023e-01 0.262 0.793550
## saving_acctA64 -8.765e-01 5.570e-01 -1.573 0.115607
## saving_acctA65 -1.057e+00 2.927e-01 -3.613 0.000303 *
## installment_rate 3.240e-01 9.618e-02 3.368 0.000756 *
## sexA92 1.701e-01 4.374e-01 0.389 0.697452
## sexA93 -4.627e-01 4.277e-01 -1.082 0.279346
## sexA94 -5.573e-02 5.127e-01 -0.109 0.913440
## other_debtorA102 6.377e-01 4.276e-01 1.491 0.135880
## other_debtorA103 -1.177e+00 4.610e-01 -2.553 0.010671 *
## age -1.534e-02 9.226e-03 -1.663 0.096272 .
## other_installA142 -4.377e-01 4.941e-01 -0.886 0.375685
## other_installA143 -8.460e-01 2.578e-01 -3.281 0.001033 **
## telephoneA192 -3.358e-01 2.062e-01 -1.629 0.103378
## foreignA202 -1.249e+00 6.531e-01 -1.912 0.055876 .
## ---
## Signif. codes: 0 '*' 0.001 '*' 0.01 '' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 977.38 on 799 degrees of freedom
## Residual deviance: 724.74 on 766 degrees of freedom
## AIC: 792.74
##
## Number of Fisher Scoring iterations: 5