Day 3 Hoop Hayden Case Roles

...

[pic 2]

Problem 3: Find an appropriate cost estimation model.

We performed three regression analyses.

For the first one we used inventory spoilage (dependent variable) and square footage (independent variable) as the variables as we want to investigate how the inventory spoilage depends on the square footage. We then try to explain the effects that the independent variable has on the dependent variable.

Regression Statistics

Multiple R

0.918755217

R Square

0.844111148

Adjusted R Square

0.832119698

Standard Error

556.4825619

Observations

15

The observations lead to an R² of 84.4%. This means that 84.4% of the inventory spoilage can be explained by the square footage. This is quite a high number and indicates that square footage is very significant when determining inventory spoilage. The causality can be questioned however as we only have a very limited number of observations. We can see the regression line below.

[pic 3]

We will now move on to our second regression.

For the second one we used inventory spoilage (dependent variable) and number of employees (independent variable) as the variables as we want to investigate how the inventory spoilage depends on the number of employees. We then try to explain the effects that the independent variable has on the dependent variable.

Regression Statistics

Multiple R

0.284303361

R Square

0.080828401

Adjusted R Square

0.010122893

Standard Error

1351.270603

Observations

15

The observations lead to an R² of 8.1%. This means that 8.1% of the inventory spoilage can be explained by the number of employees. It therefore seems that the number of employees is not very significant when determining the inventory spoilage. This can also be seen in the regression line below that does not fit the points very well.

[pic 4]

We will now move on to our third regression.

For the third one we performed a multiple regression using inventory spoilage (dependent variable) and two independent variables (square footage and number of employees).

Regression Statistics

Multiple R

0.950895

R Square

0.904201

Adjusted R Square

0.888234

Standard Error

454.0532

Observations

15

The observations lead to an R² of 90.4%. This means that 90.4% of the inventory spoilage can be explained by the two independent variables. This number is higher than the individual regression analyses as we are using two independent variables for our analysis here. Even though the number of employees alone is not very significant, combined with the square footage it yields to a higher R² than just performing a regression based on square footage alone.