1. Scenario: A MLR dataset contains healthcare related data of 100 cities in the U.S. List of variables in dataset include:

• Death rate 1000 residents
• Doctor availability per 100,000 residents
• Hospital availability per 100,000 residents
• Annual per capita income (in $1000)
• Population density per square mile
  1. List the predictors that have a significant relationship with the response

A multiple linear regression model was conducted on the dataset with the following set up:

Response: Death rate

Predictors: Doctor availability, Hospital availability, Annual per capita income, Population density per square mile

The output of the regression model is listed below:

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 12.2662552 2.0201467 6.072 1.95e-07 ***

`Doctor Availability` -0.163916 0.0069336 -3.066 0.0117

`Hospital Availability` -0.015837 0.0007219 -0.809 0.0928

`Annual Income` 0.3302302 0.2345518 1.408 0.1656

`Population Density` 0.014629 0.0048868 1.936 0.0427 .

---

Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 1.601 on 95 degrees of freedom

Multiple R-squared:0.1437, Adjusted R-squared: 0.1429

F-statistic: 3.014 on 4 and 95 DF, p-value: < 0.05

Answer the following questions based on the output of the regression model (20 points)

1. For each of the significant predictors, describe the relationship between the response and the predictor (Is the relationship positive or negative? How strong is it?). Explain your answer.
2. Evaluate the overall accuracy of the regression model using quantitative measure(s) from the output.