Я использую следующую модель:
> model1 <- lrm(dependent_variable ~ var1 + var2 + var3, data = merged_dataset, na.action="na.delete", weights = balance,normwt=TRUE)
Warning message:
In lrm(dependent_variable ~ var1 + var2 + var3+ :
currently weights are ignored in model validation and bootstrapping lrm fits
Что означает это предупреждение и как мне от него избавиться?
РЕДАКТИРОВАТЬ:
> test <- data.frame(var1,var2,var3,dependent_variable,balance)
> head(test)
var1 var2 var3 dependent_variable balance
1 -1 0 1 0 159941.89
2 1 0 1 0 138374.47
3 -1 0 1 0 157445.20
4 -1 1 0 0 44593.68
5 1 -1 -1 0 176284.48
6 -1 -1 -1 0 406905.41
> model_7 <- lrm(dependent_variable ~ var1 + var2 + var3, data = test, na.action="na.delete", weights = balance,normwt=TRUE)
Warning message:
In lrm(dependent_variable ~ var1 + var2 + var3, data = test, na.action = "na.delete", :
currently weights are ignored in model validation and bootstrapping lrm fits
> blackrock_model_7
Logistic Regression Model
lrm(formula = dependent_variable ~ var1 + var2 + var3, data = test,
na.action = "na.delete", weights = balance, normwt = TRUE)
Sum of Weights by Response Category
0 1
10049.396 2012.604
Model Likelihood Discrimination Rank Discrim.
Ratio Test Indexes Indexes
Obs 12062 LR chi2 188.52 R2 0.026 C 0.484
0 6286 d.f. 3 g 0.372 Dxy -0.031
1 5776 Pr(> chi2) <0.0001 gr 1.451 gamma -0.038
Sum of weights12062 gp 0.060 tau-a -0.015
max |deriv| 6e-10 Brier 0.137
Coef S.E. Wald Z Pr(>|Z|)
Intercept -1.3538 0.0363 -37.32 <0.0001
var1 -0.2324 0.0363 -6.41 <0.0001
var2 -0.2649 0.0350 -7.56 <0.0001
var3 -0.0640 0.0454 -1.41 0.1586