CODEWITHSUNDEEP
rlinear-regressionmissing-data
Debutant
Debutant

Reputation: 357

How to handle missing values in linear regression?

I have a data frame with 60 variables and all variables have missing values in a way that none of the lines are complete:

complete.cases(data)
  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
 [28] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
 [55] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
 [82] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE
[109] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[136] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE

So I guess I cannot use linear regression to impute missing values. Any idea how I can handle them so that I can perform my linear regression?

Upvotes: 0

Views: 1668

Answers (1)

oszkar
oszkar

Reputation: 992

Handling missing values may include:

  1. It's the best to omit variables for which most observations are missing.

  2. Omitting the rows/observations/cases with missing values. This strategy is referenced as listwise deletion or complete case analysis. This is a possibility, if the type of missingness is MCAR (Missing Completely At Random), and there are still large enough sample after the deletions.

  3. Different imputation techniques: mean/median/mode substitution, regression inputation, expectation-maximization (EM), multiple inputation, etc.

Upvotes: 1

Related Questions