The data file for this question is a diamond dataset availab…
The data file for this question is a diamond dataset available from the Seaborn website. To load this data, run the following: data = sns.load_dataset(‘diamonds’) (please note that the library has to be imported first) Create test and training datasets using the carat, table, and depth columns as the independent variables and the price as the dependent variable. (The x, y, and z columns contain information that’s related to the table and depth columns, so it’s not necessary to use those columns.) The test dataset should consist of 30% of the total dataset, and you should specify a value for the random_state parameter. ( 10 pts) Create and fit a multiple linear regression model. ( 5 pts) Find the MSE accuracy of the model with the test dataset. ( 5 pts) Create a DataFrame that shows the actual price and the predicted price. Then, display the first five rows of data to see how close the predicted prices are. (use the test set only!) (5 pts) Calculate the residuals (residual is the difference between the actual y and the predicted y) and store the results in a new column in the DataFrame you created in the previous question. Then, display the first five rows of the dataframe. (5 pts) Plot a density plot of the residuals and comment on the shape of the distribution. (5 pts) Repeat parts B and C and fit a quadratic polynomial regression. Which model is more accurate? (15 pts)
Read Details