Lecture-7: Backward Elimination Method for Feature Selection in Machine Learning by using python
This is the Video about to select the most significant features by using Backward Elimination method.
hear, 'P' value is called "significance value=0.05".
the main difference between the linear model and the statistical model is intercept(b) in y=mx+b. here, intercept is also called bias.
1. linear model is used for predicting the target variable based on test data.
2. statistical model decides the most significant independent variable and this model will never assign bias(intercept) to existing data.
3. whenever we call linear model, by default system will add bias (intercept=b) but in the statistical model, the system will not add the bias.
4. By doing Backward Elimination, we are not making a model for prediction but we try to find the relationship between variables.
5. For doing prediction, use the linear regression model.