correlation coefficients for a matrix that includes NaNs

1 次查看(过去 30 天)
I am trying to determine correlation coefficients and p values for different parameters of timeseries data. I have a matrix of five variables (each variable is a column and rows are times). I don't have data for every parameters at every timestamp so there are NaNs during the no value periods. Is there a function that will determine correlation coefficients for a dataset that includes NaNs? Right now I am using corrcoef and it gives me matrices for the pvalues and coefficients that are all 5 by 5 and filled with all NaNs. Any ideas?
Right now I am doing this: [R, P] = corrcoef(PK6mab) using the matrix in the mat file attached

采纳的回答

David Hill
David Hill 2019-9-20
[R,P]=corrcoef(rmmissing(PK6mab));
If you can tolerate deleting the entire rows of any missed data.

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Descriptive Statistics 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by