-
Linear Regression – Assumption – 2 (How To Detect & Avoid Multicollinearity ?)
Linear Regression – Assumption- 2 (How To Detect & Avoid Multicollinearity ?) Table Of Contents: How To Detect Multicollinearity In The Dataset ? Correlation Matrix. Variance Inflection Factor Model Behavior Observation. How To Avoid Multicollinearity In The Dataset ? Remove One of the Correlated Variables Use Principal Component Analysis (PCA) Use Regularization Techniques (Ridge/Lasso) (1) How To Detect Multicollinearity In The Dataset? Method – 1: Correlation Matrix (Pearson correlation) We will use Pearson ‘r’ Correlation Coefficient to find the correlation between two variable. import seaborn as sns import matplotlib.pyplot as plt # Load dataset tips = sns.load_dataset("tips") # Compute the
