1

I am trying to apply adf.test to all columns of a dataframe. Some columns have NA values. I would like to store the output in a vector/list/dataframe.

H<-data.frame(replicate(10,sample(0:20,10,rep=TRUE))) 
H[c(2,3,7,9),9]<-NA
H[c(1,4,8),2]<-NA
H[c(1,2,3,4,8),4]<-NA

I get errors as below

oi<-adf.test(H,alternative="stationary")
Error in adf.test(H, alternative = "stationary") : 
  x is not a vector or univariate time series

lapply(na.omit(H),adf.test)
Error in res.sum$coefficients[2, 1] : subscript out of bounds

Package: tseries

Help is appreciated

oivemaria
  • 453
  • 4
  • 20

1 Answers1

1

You could try

lapply(H, function(x) adf.test(x[!is.na(x)], 
           alternative='stationary', k=0))
akrun
  • 874,273
  • 37
  • 540
  • 662