Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

Please go into detail about how you solved! Thankyou A researcher studying Total

ID: 3134496 • Letter: P

Question

Please go into detail about how you solved! Thankyou

A researcher studying Total Serious Crimes (TSC) obtained a data set from the U.S. Bureau of Census (see the Minitab data file named TSC.mtw). The data set provides information for 109 Standard Metropolitan Statistical Areas (SMSAs) in the United States. The data set contains five variables: TP, PPCC, PP65, PH, and TPI. Descriptions of the variables can be found in the table below. The researcher believes that all these five variables can be used to predict TSC.                                         14 points

Variable

Definition

Description

TP

Total population

Estimated population (in thousands)

PPCC

Percent of population in central cities

Percent of SMSA population in central city or cities.

PP65

Percent of population 65 or older

Percent of SMSA population 65 years old or older

PH

Percent high school graduates

Percent of adult population who completed 12 or more years of school

TPI

Total personal income

Total current income received by residents of the SMSA from all sources (in millions of dollars)

TSC

Total serious crimes

Total number of serious crimes

Perform a regression analysis. Report the initial model.   2 points

Does the initial model contain insignificant predictors? What are the two main reasons why a model contains insignificant predictors?       3 points

Is the initial model plagued with multi-collinearity? How do you know?   3 points

How would you deal with the multi-collinearity issue? 2 points

After dealing with the multi-collinearity issue, are the remaining predictors all significant?                                                                        1 point

What is your final regression model? 3 points

Variable

Definition

Description

TP

Total population

Estimated population (in thousands)

PPCC

Percent of population in central cities

Percent of SMSA population in central city or cities.

PP65

Percent of population 65 or older

Percent of SMSA population 65 years old or older

PH

Percent high school graduates

Percent of adult population who completed 12 or more years of school

TPI

Total personal income

Total current income received by residents of the SMSA from all sources (in millions of dollars)

TSC

Total serious crimes

Total number of serious crimes

Explanation / Answer

From Minitab Using the following path we get the data as below.

Stat-Regression-Regression


The regression equation is
TP = 154 + 0.360 PPCC - 0.79 PP65 - 2.32 PH + 0.127 TPI + 0.00236 TSC


Predictor Coef SE Coef T P
Constant 153.79 37.79 4.07 0.000
PPCC 0.3597 0.2158 1.67 0.099
PP65 -0.785 2.294 -0.34 0.733
PH -2.3225 0.4673 -4.97 0.000
TPI 0.127202 0.005979 21.27 0.000
TSC 0.0023648 0.0006170 3.83 0.000

the initial model contain insignificant predictors

P value for PPCC and PP65 is higher than 0.05 hence at 5% level of significance we can say that we can

remove PPCC and PP65 from our model

Also here R-Sq(adj) = 97.4%

Now leaving the two variables the predicted model is


The regression equation is
TP = 155 - 2.20 PH + 0.124 TPI + 0.00273 TSC


Predictor Coef SE Coef T P
Constant 154.87 25.89 5.98 0.000
PH -2.1958 0.4624 -4.75 0.000
TPI 0.124133 0.005436 22.83 0.000
TSC 0.0027304 0.0005510 4.96 0.000

From the p value we can say that after dealing with the multi-collinearity issue, the remaining predictors all significant.   

R-Sq(adj) = 97.4%

hence there is no change in adjusted R square

R square defines the percentage of variation in the response reflected by the model.