Compute Covariance Matrix for real and foretasted data error

1 view (last 30 days)
Dear All,
I have one year hourly real and forecasted energy price data.I need to compute covariance matrix for error.I want to put this matrix for daily operation,which means this matrix must be 24*24.
Thank you for your guidance.

Answers (1)

Matt J
Matt J on 29 Sep 2013
Use the COV command.
  2 Comments
mehrdad
mehrdad on 1 Oct 2013
Dear Matt,
Thank you very much for your guidance,i have calculated hourly error for 365*24,and used cov command for error variance calculation,given me 365*24 matrix,i need to do this calculation for 24 hours,means ,cov error matrix must be 24*24,I have attached my data in excel,please if possible,guide and help me.
Many thanks,
Matt J
Matt J on 1 Oct 2013
mehrdad,
Applying the COV command to a 365x24 matrix should have given you a 24x24 covariance matrix as in the following short example
>> A=rand(365,24); B=cov(A); whos B
Name Size Bytes Class Attributes
B 24x24 4608 double

Sign in to comment.

Categories

Find more on Time Series in Help Center and File Exchange

Tags

No tags entered yet.

Community Treasure Hunt

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

Start Hunting!