Regresja logistyczna: Przykładowe dane: y <- c(5,8,15,25) N <- c(100,100,100,100) x <- factor(c(1:4))
Komenda glm(cbind(y,N-y)~x,family=binomial) dopasowuje do danych model regresji logistycznej. Wynik estymacji (w kolumnie Estimate jest logarytm OR czyli ilorazu szans): Call: glm(formula = cbind(y, N - y) ~ x, family = binomial)
Deviance Residuals: 0 0 0 0
Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) -2.9444 0.4588 -6.417 1.39e-10 *** x2 0.5021 0.5886 0.853 0.393606 x3 1.2098 0.5375 2.251 0.024407 * x4 1.8458 0.5137 3.593 0.000326 ***
Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 2.0424e+01 on 3 degrees of freedom Residual deviance: -2.2871e-14 on 0 degrees of freedom AIC: 24.455
Number of Fisher Scoring iterations: 3