Observation deleted due to missingness in R

Question

I am busy with a regression model in R and I have about 16 000 observations. One of these observations causes me to get the following error message:

# (1 observation deleted due to missingness)

Is there a way in R to identify this one observation?

You can check if you have missing values in your data with `sum(is.na(variable))`. With `which(is.na(variable))` you can actually identify the index (indices) of your missing observation(s). — Peter Diakumis, Feb 10 '15 at 10:17
Try `which(apply(data, 1, function(x) any(is.na(x))))` to identify the row. — lukeA, Feb 10 '15 at 10:23

score 12 · Accepted Answer · answered Feb 10 '15 at 10:48

12

If your data is in a data.frame x, and each row corresponds to an observation, then the way to go about this is to identify complete cases via complete.cases(x). Conversely, to find missing values in an observation, do ! complete.cases(x). To find out which observation contains missing values, do

which(! complete.cases(x))

answered Feb 10 '15 at 10:48

Konrad Rudolph

530,221
131
937
1,214

complete.cases worked. to identify the observation. Apparently it pulled the empty row from the csv file that was after the last observation. – Jason Samuels Feb 10 '15 at 13:04

score 2 · Answer 2 · edited May 27 '22 at 02:04

2

1 - Give your model a name like for example MyModel:

MyModel <- glm (......)

2 - Get the deleted observations rank by na.action function:

na.action(MyModel)

Results are duplicated in the display

edited May 27 '22 at 02:04

jpsmith

11,023
5
15
36

answered May 27 '22 at 01:58

Urodoc

21
2

score 0 · Answer 3 · answered Feb 10 '15 at 10:26

0

There is MWE as well as a solution in this web page : https://stat.ethz.ch/pipermail/r-help/2010-February/227526.html

which(is.na(variable)) as commented by @PeterDee seems to be indeed the solution

answered Feb 10 '15 at 10:26

ClementWalter

4,814
1
32
54

Observation deleted due to missingness in R

3 Answers3

Linked