MATLAB Answers

How do I replace missing data with other values?

조회 수: 11(최근 30일)
Robert Earles
Robert Earles 2019년 4월 15일
댓글: Robert Earles 2019년 4월 15일
Hi guys,
I have some radiation data taken every 30 mins I’m trying to use, the dataset occasionally has some missing values, labelled ‘NaN’.
I have a series of equations which can be used to estimate the radiation (won’t mention them here as they’re not relevant to the problem) in case of missing collected data. I want to use the estimated values to replace any ‘NaN’ encountered.
My raw radiation data is called ‘radi’.
My code so far is:
if isempty(radi)
% LONG SERIES OF EQUATIONS
Rn=%some terms classified in equations;
else
Rn=radi;
end
This hasn’t worked however and Rn values are just being taken as radi values, and when radi shows ‘NaN’, Rn just comes out with no value on a graph, instead of taking the estimated Rn I want it to take.
I have tried isnan instead of isempty and this creates a new issue to do with vector sizes.
Long story short, is there a better method to replace my missing data (NaN) with the equations so that my radiation data set is full?
Thanks, Rob

채택된 답변

Catalytic
Catalytic 2019년 4월 15일
편집: Catalytic 2019년 4월 15일
isnan is the correct way to detect and replace nans.
Rn=radi;
Rn(isnan(radi))= ---- something else ----
  댓글 수: 1
Robert Earles
Robert Earles 2019년 4월 15일
Will give this a go, thank you!

댓글을 달려면 로그인하십시오.

추가 답변(0개)

Community Treasure Hunt

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

Start Hunting!

Translated by