Math533 Project C Essay

858 Words4 Pages
Course Project C The scatterplot shows there is a moderate positive linear relationship between credit balance and size. As the size of the household increases, the credit balance also increases. The equation of the “best fit line” (regression equation) is Credit Balance($) = 2591 + 403 Size. For every 1 that size increases the credit balance will increase by 403. The coefficient of correlation of Credit Balance($) and Size is 0.752. Since the coefficient of correlation is positive, there is a positive linear relationship between credit balance and size of household. The coefficient of determination is .562 or 56.2%. This means that 56.2% of the total variation in Credit Balance and size can be explained by their linear relationship. The hypothesis test would be Ho: M= 3.42; Ha: M doesn’t = 3.42. Using a two-tailed test with a = .05, I find that z = 13.91 which is in the rejection region of z < -1.645 or z > 1.645. The p-value using the p-value calculator = 0.0001 which is statistically significant. With the above results, I would reject the null hypothesis because the z test statistic falls within the rejection region and the p-value is less than a. There is significant evidence that increasing the family size will also increase the credit balance. According to the above findings, I can say that using household size to determine credit balance is 56.2%accurate when used alone without any other variables. In my opinion we need other variables to determine credit balance more accurately. The 95% confidence interval for the population slope is (3.35; 3.49). This means that we are 95% confident that the mean family size of our customers will fall between 3.35 and 3.49. The 95% confidence interval of the average credit balance for a family size of 5 is (4368.20; 4846.90). This means that we can be 95% confident that for a family size of 5 the mean
Open Document