Answer to part
a)
enter data of x and y in excel
click on data tab
in it click on data analysis
in it select regression
click ok
The following window appears on screen:
.
In the regression window :
enter the range of Y values
enter the range of X values
check the box for labels
check the box for constant zero (this implies the regression equation will have no constant value)
click ok
the following output is obtained:
The model is: Y = 51.17322 x
.
Part b)
The model test values are: F = 24146.71 , with an extremely low significance value 3.29 x 10^-15
Since the significance value < 0.05 , this implies there is evidence that x contributes information for prediction of y
.
Part c)
In this scenario we repeat all the steps used in part, and do not check the box for constant zero
as follows:
.
enter the range of data for y
enter the range of data for x
tick mark the check box of label
click ok
the following output appears
.
Now again since the P value = 1.89 x 10^-13 < 0.05 , this implies the value of x contributes significant information to predict the value of Y
.
Part d)
The model with the highest R square value must be recommended
Since both the models are significant, the r square value will tell us the percent of variation in Y that can be explained by the model. Hence we compare the R square values for both the models:
R square value for model 1 = 0.999627
R square value for model 2 = 0.999089
Hence model 1 has higher value of R square, thus one must go for model # 1
7. An electric utility company must develop a reliable model for projecting the number of resi- d...