PCA in Regression analysis

r
machine_learning

#1

Hi Experts,

Please point me to some example how pca can be called in regression Analysis?

insights much appreciated.

regards,
tony


#2

PCA Will tell you the important variables from the dataset and it will help you to create a model in regression analysis with less number but important features

useful links - http://stats.stackexchange.com/questions/27300/using-principal-component-analysis-pca-for-feature-selection

http://www.stats.uwo.ca/faculty/braun/ss3850/notes/sas10.pdf


#3

Hi,HUNAIDKHAN2000,

Thanks for you valuable inputs.
I am a Newbie and looking for hands on PCA calling in regression Analysis with some practical data.
many practical example only explains about PCA analysis. But not its use case in a Reression Analysis

Please point me if u find any.

Thanks,
tony


#4

Hi @tillutony.

So firstly, to to understand PCA refer to https://www.analyticsvidhya.com/blog/2016/03/practical-guide-principal-component-analysis-python/ in addition to the other resources mentioned by @HUNAIDKHAN2000 . If you are havind trouble handling the dimentionality of your data, you should opt to go for PCA.

Secondly, for regression analysis, you can simply pick up first N Principal components to build a model on. To find the best no. of principal components, you’ll have to see the CV scores for each N. I would suggest to go for binary search instead of brute force approach for deciding N.

Thirdly, from the perspective of model interpretation, I’m sorry you won’t be able to interpret the model based on the original components.