0

I want to explore various classification methods on a data set with many missing data. I tried imputing with cart method of mice since I was getting a singular error for the default method. My problem is that, after imputing, I still have missing data. What I did:

imputed_Data <- mice(stock, m=1, maxit = 5, method="cart", seed = 328)
data_det <- complete(imputed_Data)

Any insights greatly appreciated!

Kkk
  • 59
  • 7
  • It's easier to help you if you include a simple [reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) with sample input and desired output that can be used to test and verify possible solutions. How exactly is your `stock` data formatted? How many columns have missing data? – MrFlick Apr 02 '21 at 22:57
  • All columns have missing data, 223 columns (features) and 4000+ rows (observations). If I just exclude all columns that have missing data I get an empty data set. All my values are numeric. – Kkk Apr 03 '21 at 00:26
  • Also the imputed data have much fewer columns with missing but still I have missing values – Kkk Apr 03 '21 at 00:48

1 Answers1

2

My issue was solved with setting ridge = 0.0001 and threshold=1.1 and I got the complete dataset

Kkk
  • 59
  • 7