| compute_weights {CARRoT} | R Documentation |
Weights of predictors
Description
Function which computes the weight of each predictor according to the rules of thumb and outputs it into corresponding array
Usage
compute_weights(vari_col, vari)
Arguments
vari_col |
number of predictors |
vari |
set of predictors |
Details
Continuous or categorical numerical variable with more then 5 categories has weight 1, otherwise it has weight n-1 where n is the number of categories
Value
Returns an array of weights of the size vari_col
References
Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR (1996). “A simulation study of the number of events per variable in logistic regression analysis.” Journal of Clinical Epidemiology, 49(12), 1373-1379. ISSN 0895-4356, doi:10.1016/S0895-4356(96)00236-3, http://dx.doi.org/10.1016/S0895-4356(96)00236-3.
Rhemtulla M, Brosseau-Liard PÉ, Savalei V (2012). “When can categorical variables be treated as continuous?: A comparison of robust continuous and categorical SEM estimation methods under suboptimal conditions.” Psychological Methods, 17(3), 354-373. doi:10.1037/a0029315.
Examples
#creating data-set with for variables
a<-matrix(NA,nrow=100,ncol=4)
#binary variable
a[,1]=rbinom(100,1,0.3)
#continuous variable
a[,2]=runif(100,0,1)
#categorical numeric with les than 5 categories
a[,3]=t(rmultinom(100,1,c(0.2,0.3,0.5)))%*%c(1,2,3)
#categorical numeric with 5 categories
a[,4]=t(rmultinom(100,1,c(0.2,0.3,0.3,0.1,0.1)))%*%c(1,2,3,4,5)
#running the function
compute_weights(4,a)