google collaborators probabilities

CE 316: Applied Probability and Statistics in CEE

Homework 8

Due: December 5 at 10:00 PM

Problem 1:

Upload the Neth.CSV data file to google colab. Estimate a linear regression model of number of weekly trips per household as a function of the remaining variables (to the extent possible). In the final model identify which variables contribute to an increase in number of weekly trips and which variables contribute to a decrease in the number of weekly trips.

In estimating a linear regression model these are the steps I want you to do:

• Run multiple linear regression by including all variables

• Check for multi-collinearity and remove appropriate variables.

• Check if all variables are significant and remove the variables which are not significant one by one.

• Report the R2 and Adjusted R2 of the final model

• Check if the mean of the residuals is close to zero. Comment on the mean.

• Plot the histogram of the standardized residuals. Comment on whether the standardized residuals look normal.

• Check for outliers.

• Plot the residuals vs fitted values and comment.

The variable definitions are:

• HHSIZE household size

• NCAR number of cars in household

• HEMPSTS number of workers in household

• HSTUDEN number of students in household

• HTTRPS number of weekly trips per household

• NUCHLT12 number of children < 12 years in household

• CITY household residence in city (dummy variable)

• SUBURB household residence in suburb (dummy variable)

• RURAL household residence in rural area (dummy variable)

• INCOME continuous household income value

• NUCHGT12 number of children >= 12 yrs in household

8-1