I have a dataset showing monthly stock index returns for the last twenty years across 7 regions. Each region has 6 stock indices( both growth and value stock indices for small cap, standard cap and large cap companies), giving me a total of 42 monthly stock return time series.

I want to investigate the correlation between some of the growth and value returns within the same regions, across different regions and across different company sizes, as well as comparing it to GDP for the region (a statistic I will download from the World Bank).

Do I need to normalise my data before I calculate the correlation coefficients? Or does this assumption not hold for the data I am using for some reason?

  • 3
    $\begingroup$ Pearson correlation coefficient is the same whether you standardize the variables or not. $\endgroup$ – ttnphns Apr 26 at 17:16

Correlation doesn't need you to standardize first.

Correlation is the covariance divided by the product of the standard deviations of each variable. So, in its computation, the standardization is done by definition.

Covariance of standardized variables equals the correlation.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.