TRI - using MIRT

Asked

Viewed 176 times

2

I am applying the TRI (Item Response Theory) with the package mirt, in Rstudio. However, when applying the mirt, processing stops when it reaches 500 iterations. Is it possible to increase the number of iterations? If so, how?

r

2 answers

3

The maximum number of iterations is documented in the package help page, function mirt.

Of help("mirt").

Technical 

    a list containing lower level technical parameters for  
    estimation. May be:  

NCYCLES maximum number of EM or MH-RM cycles;  
        defaults are 500 and 2000

Translation Google Translate.

Technical 

   uma lista contendo parâmetros técnicos de nível inferior  
   para estimação. Talvez:  

NCYCLES  número máximo de ciclos EM ou MH-RM;  
   os padrões são 500 e 2000

Testing with one of the examples that does not converge with the number of standard iterations, the example with the dataset SAT12, you can see it works.

library(mirt)

data(SAT12)
data <- key2binary(SAT12,
                   key = c(1,4,5,2,3,1,2,1,3,1,2,4,2,1,5,3,4,4,1,4,3,3,4,1,3,5,1,3,1,5,4,5))

mod2 <- mirt(data, 2, optimizer = 'NR')
#Iteration: 500, Log-Lik: -9441.963, Max-Change: 0.00012
#EM cycles terminated after 500 iterations.

mod2.b <- mirt(data, 2, optimizer = 'NR', 
             technical = list(NCYCLES = 5000))
#Iteration: 1065, Log-Lik: -9441.950, Max-Change: 0.00010

In this second case went up to the iteration 1065 and has reached tolerance 0.0001.

1

Bringing the translation of the answer I found in a Google Groups:

"It is on the technical entry list, called NCYCLES:

mod <- mirt(Science, 1, technical = list(NCYCLES = 2000))

source:Google Groups

Browser other questions tagged r

You are not signed in. Login or sign up in order to post.