Linear regression

#1

sir, lets assume we have multiple feature in X_train but 1 feature is highly correlated with Y-train and others are very less correlated so should we go for simple linear reg with that 1 feature or we should take alll feature ie,multiVar linear reg

#2

hey @Ashu1318 ,
we can’t guarantee that simple linear regression will work or not.
So in this case , the most important thing will be experiment.

maybe the independent features , can have a correlation between them which can be used to improve the model and score better.

#3

okay and whats role of normalisation ?? is it used when data is more spreaded or any time we can use it??

#4

normalization is basically converting your data in such way that the mean value becomes zero , convert it more closer or similar to gaussian distribution.

It basically used based on the model used , like the effect of normalization on tree models is not that much , but on linear models it very huge.
So , you need to decide by your when to use and when to not.

#5

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.