In estimating the regression in problem #2, you are also concerned that the t-statistics may be inflated because of the presence of conditional heteroscedasticity. There are 214 observations and 3 independent variables.

A researcher is investigating possible explanations for deaths in traffic accidents. He examined data from 2000 for each of the 52 cities randomly selected in the US. The data included information on the followingvariables: Deaths: The number of deaths in traffic accidents per city and Income: The median income per cityAs part of his study, he ran the following simple linear regression model as pictured :  Question:  Based on the above results, the researcher tested the hypotheses ( Null: B1=0 versus Alternative: B1 not equal to 0) using T test. What do we know about the test statistic of the test, what is the approximate p-value, and value of Rsquared? And based on your result, what is your conclusion? Show your work for full credit.

Where we observe multicollinearity in a multiple regression analysis, two of our independent variables, x-1 and x-4, are so highly correlated they’re almost indistinguishable with respect to their relationship with our dependent (y) variable. Why is that a problem?

Suppose you are to estimate a simple regression for the following population model: Y=B₁ + B₁X + µl From a population of over thousands of observations, a small number of samples were randomly selected. The following is some of the information from the randomly selected sample.