A concise, comprehensive treatment of the application of statistical regression analysis suitable for undergraduate and graduate students. Covers simple linear regression, multiple linear regression, model building, and advanced regression topics.
Answer the following questions based on Section 5.2.5: Regression pitfalls - overfitting.
Consider the following data:
Use statistical software to fit a simple linear regression model with response variable, Y, and predictor variable, X. Obtain the two-tailed p-value for the regression parameter for X. Round your answer to 4 decimal places.
Create a scatterplot with Y on the vertical axis and X on the horizontal axis and add the fitted simple linear regression line from Q1 to the plot. The plot should look like the following.True or false? The scatterplot indicates that the simple linear regression model is reasonable.
Use statistical software to fit a multiple linear regression model with response variable, Y, and predictor variables, X, X2, X3, X4, and X5. Obtain the two-tailed p-value for the regression parameter for X. Round your answer to 4 decimal places.
Create a scatterplot with Y on the vertical axis and X on the horizontal axis and add the fitted multiple linear regression line from Q3 to the plot. The plot should look like the following.True or false? There is a clear nonlinear pattern in the data that the multiple linear regression model from Q3 appropriately captures.
The fitted regression line in the plot in Q4 indicates that as X increases, first E(Y) tends to increase, then E(Y) tends to decrease, then E(Y) tends to increase again, then E(Y) tends to decrease again, then finally E(Y) tends to increase again. True or false? This pattern likely represents the true association between Y and X in the population.
Time's up
Categories:
Tags:
Comments are closed
Comments are closed