Pearson Correlation with Group By

Hello,
I have a sample csv file as below attached. For 1 user has many records. In this case user 91 has 5 and user 135 has 4 records. I need to calculate the Pearson correlation among rating and my rating columns. Ideally my output should be like below, Need to find Value 1 and Value 2.

I noted that to get this output I need to use a batch macro. I tried using some of the community answers but was unable to figure it out. If someone can help me on this greatly appreciated. It could have been easier if the feature was in built in Pearson correlation tool. Just a suggestion.
Thank you.

test.csv

Macros

Predictive Analysis

Accepted answers

phottovy

Here is the same workflow with a simple batch macro that gives you the results. I left it fairly basic but you could easily add options that let you select which columns to use.

Pearson Correlation Group By.yxzp

All comments

phottovy

Hi @suresh_abeyweera

I'd have to refresh my memory on the difference but you can use the Spearman Correlation tool with a group by. I will also put together a quick macro to get you the same results for the Pearson Correlation

Spearman Correlation.yxmd

phottovy

Here is the same workflow with a simple batch macro that gives you the results. I left it fairly basic but you could easily add options that let you select which columns to use.

Pearson Correlation Group By.yxzp

suresh_abeyweera

Thank you.

It seems that the correlation values from spearman and Pearson drastically change. May be I would like to think which to be used in my problem.(In recommendation System)...

Quick Links

This months top contributors

atcodedog05 19458

Qiu 15866

binu_acs 15708

MarqueeCrew 13708

apathetichell 13703