Correlation between multiple variables

Hi,

have an issue with the correlation for multiple variables, using pearson or others i can't find for multiple variables in a go with a group by function.

e.g

i have Revenue as column and have to check whether Cost, margin, discount have any correlation to Revenue but according to the year and location.

#correlation

#groupby

Accepted answers

ImadZidan

Hello @nidah5 ,

@echuong1 is quiet right about the grouping feature.

Please have a look at the attached and it visualised what @echuong1 was saying about creating filters and what have you.

One thing I should point out is that you need to have at least one observation to apply the PCorr.

Now, having said said the workflow should give you an idea. What I would say, you can also change the workflow to run as a macro may be an iterative or a Batch. At least you have a starting point.

Any problem, shout back.

CorrRevenue.yxzp

All comments

ImadZidan

Hello @nidah5 ,

Can you please provide some sample dummy data. It will help.

We will explore this task together.

echuong1

There is not a group by option in that tool, or any of the predictive tools. To do this on a per year and location basis, you'd have to filter for each combination. Ex: filter for location #2 and year 2018, and then do the correlation.

nidah5

Hi,

have attached a file

it has 5 variables

CUSTOMER & DATE field which should be used as group by

Revenue as Y, AND Cost & Margin as X

So i need to check for a particular CUSTOMER on given DATE, what's the correlation of revenue with cost & revenue with margin.

attached file has RAW DATA and Second CORRELATION Sheet as the the format of final product.

please let me know if this helps @ImadZidan @echuong1

SAMPLE.xlsx

Quick Links

This months top contributors

atcodedog05 19458

Qiu 15866

binu_acs 15708

MarqueeCrew 13708

apathetichell 13703