CODEWITHSUNDEEP
rlinear-regression
Samoth
Samoth

Reputation: 756

r- how to store the result of boxtidwell into a data frame/matrix

I'd like to store the result of the boxtidwell in order to write some code to achieve automation and avoid manually transform the variables. See example below:

 >boxTidwell(prestige ~ income + education, ~ type + poly(women, 2), data = Prestige) 

 ##            Score Statistic   p-value MLE of lambda
 ## income          -4.482406 0.0000074    -0.3476283
 ## education        0.216991 0.8282154     1.2538274 
 ## iterations =  8

Because I do not want to transform the variable manually such as the code below:

 >Prestige$income <- (Prestige$income )^(-0.3476283)

When there are many variables need to be transformed then it will be so time-consuming. I tried matrix and data.frame but both in vain. 

> box<-boxTidwell(prestige ~ income + education, ~ type + poly(women, 2),   data = Prestige)
> box<-as.data.frame(box)

#Error in as.data.frame.default(box) : 
#cannot coerce class ""boxTidwell"" to a data.frame

> box<-as.matrix(box)
> box

#Error in round(x$result, digits) : 
#non-numeric argument to mathematical function

I've searched for a while but seems in vain, many thanks for any suggestions and ideas in advance.

Upvotes: 0

Views: 93

Answers (1)

Eugene
Eugene

Reputation: 406

It's accessible in box$result, e.g. 

box$result[,"MLE of lambda"]

    income  education 
-0.3476283  1.2538274

so your code above becomes (crudely)
Prestige$income <- (Prestige$income )^box$result[1,3]

In general, to check out what's inside your object use str

Upvotes: 1

Related Questions