Math 533: Applied Managerial Statistics
Course Project Part C: Regression and Correlation Analy
Summary
After statistical analysis of data collected on a random sample of AJ Davis’ customers, several inferences can be made. The goal of this analysis was to determine the best model for predicting customer income. This knowledge is paramount in most aspects of the company including but not limited to advertising, sales and merchandising. Based on the analysis, it is determined that a model using customer credit balance and household size is the most efficient. These variables gave the most reliable predictions of customer income. At the close of the analysis, the data yielded the following equation for determining customer income, income== - 1.90 + 0.0173 CREDIT BALANCE($) - 5.30 SIZE - 0.390 YEARS. Though, the number of years a customer has live in their home did not show high effectiveness in predicting income, it increased the overall fit of the model and has been included in the equation. We can be 95% confident in the data obtained through the use of this formula.

Appendix

1.
2. Regression Analysis: INCOME($1000) versus CREDIT BALANCE($)
The regression equation is
INCOME($1000) = 4.45 + 0.00987 CREDIT BALANCE($)
Predictor Coef SE Coef T P
Constant 4.448 7.037 0.63 0.530
CREDIT BALANCE($) 0.009866 0.001728 5.71 0.000
S = 11.3247 R-Sq = 40.5% R-Sq(adj) = 39.2% Analysis of Variance
Source DF SS MS F P
Regression 1 4182.2 4182.2 32.61 0.000
Residual Error 48 6156.0 128.2
Total 49 10338.2 Unusual Observations
CREDIT
Obs BALANCE($) INCOME($1000) Fit SE Fit Residual St Resid 3 5100 32.00 54.77 2.53 -22.77 -2.06R 5 1864 31.00 22.84 3.97 8.16