Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Correlation Testing

By Jonathan Crouch | Aug 15, 2016 05:42AM CEST

I want to test the correlation between two variables, ESG scores and prices for the market as a whole.
I have two datasets, one which has ESG scores for 70 companies over 48 months. The companies are in each row and the columns are each month.
My other dataset has the prices for those companies arranged in the same way (companies in rows, prices in columns)
I want to know the overall correlation between ESG scores and prices over the entire sample period. I don't want this company by company or month by month - just one number which describes the correlation between the two variables for the entire market over the entire sample period. Is this possible? How do I go about this?




By Paulin | Aug 16, 2016 06:30PM CEST | XLSTAT Agent

Dear Jonathan,

The canonical-correlation analysis (CCA) seems a suitable method in your case for cross-covariance matrices analysis. If you have two randoms vectors X and Y, the analysis will find linear combinations of the Xi and Yj which have maximum correlation.

The CCA feature is available under Advanced Features>Multiblock Data Analysis>CCorA


This question has received the maximum number of answers.

Contact Us
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
Invalid characters found