Question: Possible methodologies for missing value imputation of continuous variables in a data frame in R
0
gravatar for svlachavas
3.9 years ago by
svlachavas700
Greece/Athens/National Hellenic Research Foundation
svlachavas700 wrote:

Dear All,

i have a data frame in R of 8 continuous variables in the rows and 60 paired observations in the columns. I want to use this data frame in a subsequent analysis regarding machine learning, but unfortunatery in 3 samples the values from both the 8 variables are missing(not zero, totally missing). Thus, because if i exclude these samples from any downstream analysis, i also have to exclude their paired observations, and in totally loose 6 samples-which might seem not significant but i could loose information. Thus, i would like to ask if there is an package in R with an appropriate methodology of missing value imputation, to implement it in this current purpose. Im aware of the R package impute with k nearest neighbors, but as far as i know it is intended for gene expression data. Moreover, after searching i found also the R package MICE (https://cran.r-project.org/web/packages/mice/mice.pdf) but i have never used it. 

Any ideas or suggestions on how could i deal with this issue ??

 

ADD COMMENTlink written 3.9 years ago by svlachavas700
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 392 users visited in the last hour