In: Statistics and Probability
Charity # | Charity | Total Expenses ($) | Administrative Expenses (%) | Fundraising Expenses (%) | Program Expenses (%) |
1 | American Red Cross | 3352089148 | 3.8 | 3.9 | 92.0 |
2 | World Vision | 1208111251 | 4.0 | 7.5 | 88.3 |
3 | Smithsonian Institution | 1081275619 | 23.5 | 2.5 | 73.8 |
4 | Food For the Poor | 1049984888 | 0.6 | 2.6 | 96.8 |
5 | American Cancer Society | 1007342150 | 6.0 | 22.3 | 71.4 |
6 | Volunteers of America | 931512538 | 8.5 | 1.8 | 89.2 |
7 | Dana-Farber Cancer Institute | 876227147 | 13.1 | 1.5 | 85.3 |
8 | AmeriCares | 858665385 | 0.3 | 0.8 | 98.9 |
9 | ALSAC - St. Jude Children's Research Hospital | 830079269 | 9.6 | 17.0 | 73.5 |
10 | City of Hope | 734387170 | 13.4 | 2.9 | 83.0 |
1. Develop an estimated regression equation that could be used to predict the program expenses (%) given fundraising expenses (%)
Program Expenses % (pred) = ? + ? Fundraising Expenses (%) up to 2 decimals
Firstly, try to plot a scatterplot between these two to find out the trend.
It shows the linear relationship between the two variables, hence linear regression can be applied here.
We will be applying the Linear regression model here, it can be done by using the function LINEST(y_value, x_value, TRUE, TRUE) where y_values contain values of Program Expenses here and x_values have Fundraising Expenses values.
Select 5 rows and 2 columns and then write the formula in the first cell and after that, press Shift + Ctrl + Enter.
The equation comes out to be -
program expenses = 90.9 - 0.90*Fundraising Expenses