Error when using the mice package

[m4henqk]

Hello,

I'm encountering an error while working on my R code, and I'm hoping to get some guidance on how to resolve it. The error message I'm receiving is:

"Error in class(ff) <- "formula" : attempt to set an attribute on NULL"

I have read an excel sheet using the "readxl" package, stored it in an object named "data.pre", converted it from a tibble dataframe to a traditional dataframe to avoid errors and checked for missing data and null values using: "sum(is.na(data.pre))" and "sum(is.null(data.pre))", which have returned 60 and 0, respectively.

I am very new to R and any guidance on how I can identify and overcome this problem would help a lot.

Best regards,
Matheus

[m4henqk]

The current code is:

install.packages("readxl")
library(readxl)

data.pre = read_excel("C:/Users/josia/Documents/pre.xlsx", sheet = "Pre", range = "A1:U28", na = "NA")
sum(is.na(data.pre))

data.pre = as.data.frame(data.pre)

install.packages("mice")
library(mice)

imp_model = mice(data.pre, method = "logreg", m = 5)

Thanks in advance.

[FJCC]

Please post the output of

dput(data.pre)

Place a line with three back ticks just before and after the pasted output, like this:
```
pasted output goes here
```

[m4henqk]

I actually managed to go past that error by deleting the identifiers of all observations, it seems there were some characters that were considered NULL values. However I can't seem to proceed with multiple imputations because I get the message there is significant co-linearity. I have already excluded the data from variables that did not vary, but the error is still there.

This is the output of "dput(new.data.pos)" (I changed the object name):

> dput(new.data.pos)
structure(list(F = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 1), G = c(1, 1, 1, 0, NA, 1, 1, 1, 1, 1, 1, 1, 1, 1, 
NA, 1, NA, 1), I = c(1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, NA, 
0, 0, 0, NA, 0), K = c(1, NA, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, NA, 
NA, NA, 0, NA, 0), L = c(1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 
0, NA, NA, 0, NA, 0), M = c(1, NA, 0, 0, 0, 0, 0, 1, 1, 1, 0, 
1, 1, NA, 1, NA, NA, 1), N = c(1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 
1, 1, 1, 1, 1, 1, 1, 1), O = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
1, 0, 0, 0, 0, 0, 0, 1), P = c(1, NA, 1, 0, 1, 0, 1, 1, 1, 0, 
0, 1, NA, 1, 1, NA, 1, 0)), row.names = c(NA, -18L), class = c("tbl_df", 
"tbl", "data.frame"))

Your help is very much appreciated,
Matheus

[FJCC]

The colinearity is a different problem that probably should have a separate thread. I can't help with a solution because I have never used the mice package. I will observe that a few of your columns are nearly identical. F, I, and O are almost all zeros and G and N are almost all ones. I don't know how much of that the mice algorithm can handle or what tools it has to alleviate the problem.