Wanting to search for "multiple correlation" in my dataset. Do two or more variables together predict my target variable

user_8567437 0
Viewed 54

By Multiple Correlation I refer to this: https://en.wikipedia.org/wiki/Multiple_correlation

If I understand it right, this should give me the correlation of any two x_1, x_2 together and how they correlate to a target variable y. At least that is what I am searching for. Any other approaches would be helpful too. How do I check if there is a correlation of not one variable to my target, but of two or more together.

This is not implemented by: dataframe.corr()

Should I rather go by using Linear Regression or some simple ML model? It is important for me to see which variables are how strong in my model so to reconstruct this as a formula.

data-analysis correlation dataframe python
1 Answers

As the wikipedia page you pointed says, the Multiple correlation measure is the correlation between the value predicted by a linear regression and the input data. If you want to do any prediction you will have to compute the linear regression.

dataframe.corr you mentioned, will give the Rxx matrix mentioned by wikipedia article. You still have to compute c, to plug in the equations.

Linear regression is the best way to go, it gives you a predictor that minimizes the error, and the error you can use to decide whether you consider that a good fit.

user_127503530
Related