Linear Regression Assumptions Testing in R



Hi Team,
I am currently working on a problem of predicting the House Loan Prices in UK , The Loan prices are continuous in nature hence I was planning to use multiple linear regression as a start , however I am not sure how to test its assumptions in R , I refer few blogs in google but not able to follow much, can anyone please help with implementation of the below assumptions preferably in relatively simpler manner.I am basically trying to test below assumptions

  1. Linear relationship (between the independent and dependent variabales).
  2. Multivariate normality.
  3. No or little multicollinearity.
  4. No auto-correlation.
  5. Homoscedasticity.

Also While using COR() function to determine the correlation between the independent variables what should be the cut of value we can use to determine the strong/some** +/- correlation between them, I got 56% correlation can that be considered as strong correlation ?



Hi Arnab,

Please go through the below article and it will answer all of your questions about Linear Regression.