Plot Unit root test - Ljung-Box statistic over the residuals

Hello,
I want to build a graphic looking like that one:


Therefore I used ARIMA to model the Data. Attached you will find the code I am currently working with. But I don't know how to assign the values to the matrix and how to plot the p-values afterwards. It as well tells me:
"Error in ljung_mat[i, ] <- print(c(i, box$p.value)) :
Number of elements to be replaced is not a multiple of the replacement length "

Thank you for your help.
Kind regards
Luke

n2 = length(train)
ljung_mat <- matrix(0, nrow=(n2), ncol=1)
       for(i in 1:n2)   
         {
         box<-(Box.test(finalarima$residuals, lag=i, type="Ljung-Box")) 
         ljung_mat[I,] <- print(c(i, box$p.value))
          }

Can you try ljung_mat[i,] <- box$p.value ?

I am not sure what goes wrong on your side.
But generating some test data in variabele x this works for me :

set.seed(2020)
x = rnorm(50)
n2 = length(x)
ljung_mat <- matrix(0, nrow = (n2), ncol = 1)

for (i in 1:n2)
{
  box <- 
    Box.test(
      x,
      lag = i,
      type = "Ljung-Box")
  ljung_mat[i, ] <- box$p.value 
}

plot(ljung_mat)

Created on 2020-07-24 by the reprex package (v0.3.0)

1 Like

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

It says:
Error in [<-(*tmp*, i, , value = box$p.value) :
Indexing outside the limits

Thank you that works perfectly now! :slight_smile: