Explore BrainMass

# Regression Analysis

This content was COPIED from BrainMass.com - View the original, and get the already-completed solution here!

See attached Excel files.

Calculations

Looking for
Single Regression Models
Multiple Regression Model with ONLY statistically significant Coefficients
Explanation of data collection, SOURCE of data, regression methodology and
specific hypothesis tests F test, and T test (or p-value).

Obtain a correlation matrix
Obtain your "best" multiple regression model using the combined data
Interpret your output. Write up your explanation. Be sure to explain which independent variables are most strongly related to price (in each city #1 and all three cities #2)

? which are least strongly related to price,
? if the directions (positive, negative) of the correlations make sense,
? if there appears to be any multicollinearity,

Ho: &#946;1 = &#946; 2 = &#946; 3 = 0
Ha : &#946;1 and/or &#946; 2 and/or &#946; 3 is not equal to 0

Explanation of "best" regression model and HOW it is determined to be better
than other models. Should only include ONLY statistically significant
coefficients and the best R2 values (given the coefficients that are included)
Explanation of conclusions drawn from the results for the CAR BUYER

Selling price = Intercept + Coefficient * x1 (Age) + Coefficient * x4

What the results of the regression equation mean to the buyer of that vehicle.

For instance:
If the buyer would accept a car that was one year older, he/she would expect the selling price to be on average \$912 less. ( If - \$912 is the coefficient on the age variable and it is statistically significantly different from -0-)

If the city (dummy variable) is statistically significant then that coefficient is the difference in the selling price associated with that city (keeping the age and mileage of the car the same).

Likely either age OR mileage will be in "best" model since they are often explaining the same decrease in price. DO explain multicollinearity.