Stat460. Homework #1
Homework 1
STAT 460
Due Jan 25 by the beginning of class
Use the old faithful data used in lab 1.
You will need to use excel for this assignment and the excel file MUST be turned in with your homework.
Part 1: BY HAND (or I should say by Excel) calculate the slope and the intercept for the regression model for the data (you can double check your results using the minitab output obtained in lab. (2 pts)
Summarize your calculations here, including the value of the slope and intercept
Duration | TimeToNext | yhat | error | slope | y-intercept | 1.7 | 55 | 51.68 | 3.32 | 10.35821 | 33.96676 | 1.7 | 58 | 51.68 | 6.32 | | | 1.7 | 56 | 51.68 | 4.32 | | |
Part 2:
In Minitab re-run the regression and click on the Storage button and store the residuals and fitted values. Next, to to STAT>Descriptive statistics and obtain the standard deviation for the residuals. (1 pt)
It was done in Minitab and results were put into Excel file to use in calculating standardized value of the residuals.
Calculate the standardized value for each of the residuals. Do this by dividing each residual by its standard deviation (obtained in the descriptive statistics). You will need to first copy the column of residuals into excel, again this MUST be part of your excel file. (1 pt)
This was done in Excel (H column).
In excel sort the value of the standardized residuals from largest to smallest to identify any points that look like they may be outliers. Identify any points that you feel are outliers and discuss the values of X and Y for these points. Include the original scatterplot of the data in this discussion. (1 pt, 1 pt, 1pt)
Of these points, which points do you believe may be influential (if any?). Why or why not (make sure you discuss all points you identified in the above section). (1pt, 1pt)
Part 3: Use the Density data on the course page.
This data reflects sales as a function of population density, where the amount of sales
STAT 460
Due Jan 25 by the beginning of class
Use the old faithful data used in lab 1.
You will need to use excel for this assignment and the excel file MUST be turned in with your homework.
Part 1: BY HAND (or I should say by Excel) calculate the slope and the intercept for the regression model for the data (you can double check your results using the minitab output obtained in lab. (2 pts)
Summarize your calculations here, including the value of the slope and intercept
Duration | TimeToNext | yhat | error | slope | y-intercept | 1.7 | 55 | 51.68 | 3.32 | 10.35821 | 33.96676 | 1.7 | 58 | 51.68 | 6.32 | | | 1.7 | 56 | 51.68 | 4.32 | | |
Part 2:
In Minitab re-run the regression and click on the Storage button and store the residuals and fitted values. Next, to to STAT>Descriptive statistics and obtain the standard deviation for the residuals. (1 pt)
It was done in Minitab and results were put into Excel file to use in calculating standardized value of the residuals.
Calculate the standardized value for each of the residuals. Do this by dividing each residual by its standard deviation (obtained in the descriptive statistics). You will need to first copy the column of residuals into excel, again this MUST be part of your excel file. (1 pt)
This was done in Excel (H column).
In excel sort the value of the standardized residuals from largest to smallest to identify any points that look like they may be outliers. Identify any points that you feel are outliers and discuss the values of X and Y for these points. Include the original scatterplot of the data in this discussion. (1 pt, 1 pt, 1pt)
Of these points, which points do you believe may be influential (if any?). Why or why not (make sure you discuss all points you identified in the above section). (1pt, 1pt)
Part 3: Use the Density data on the course page.
This data reflects sales as a function of population density, where the amount of sales