Scribd
Upload
23 views2 pages

ANOVA and Regression Analysis Tutorial

The document presents a tutorial memo focused on a first order regression model, detailing tasks related to ANOVA, hypotheses testing, and the interpretation of regression coefficients. It includes specific questions regarding the necessity of predictor variables, sums of squares, and the coefficient of determination. Additionally, it provides a practical example involving hospital income and various predictors, culminating in a conclusion about the relevance of certain variables in the regression model.

Uploaded by

Lü Lä Mï Lë
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
23 views2 pages

ANOVA and Regression Analysis Tutorial

The document presents a tutorial memo focused on a first order regression model, detailing tasks related to ANOVA, hypotheses testing, and the interpretation of regression coefficients. It includes specific questions regarding the necessity of predictor variables, sums of squares, and the coefficient of determination. Additionally, it provides a practical example involving hospital income and various predictors, culminating in a conclusion about the relevance of certain variables in the regression model.

Uploaded by

Lü Lä Mï Lë
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Tutorial 5 Memo 9 September 2024

Chapter 7

Question 1

The following first order regression model is considered

𝑌𝑖 = 𝛽0 + 𝛽1 𝑋𝑖1 + 𝛽2 𝑋𝑖2 + 𝛽3 𝑋𝑖3 + 𝛽4 𝑋𝑖4 + 𝜀𝑖

1.1 Set up a template ANOVA table for this regression model, showing the decomposition of SSR.

1.2. Provide the hypotheses, test statistic formula and decision rule to answer the following research
questions.
1.2.1. Is there a statistical relationship present between the response variable and the set of
predictor variables?

1.2.2. Is the predictor variable 𝑋4 necessary in the model?

1.2.3. Are the predictor variables, 𝑋3 and 𝑋4 , necessary in the model?

1.3. Provide the formula of the coefficient of partial determination between 𝑌 and 𝑋3 , if 𝑋1 and 𝑋2 are
already in the model. Assume a value of 0.027 was determined. Interpret this value. What is the
conclusion?

Question 2

Source of Sum of Mean


Variation DF Squares Square
Regression 4 107.9 26.975
𝑆𝑆𝑅(𝑋1 ) 1 90.2
𝑆𝑆𝑅(𝑋2 | 𝑋1 ) 1 15.3
𝑆𝑆𝑅(𝑋3 |𝑋1 , 𝑋2 ) 1 1.8
𝑆𝑆𝑅(𝑋4 |𝑋1 , 𝑋2 , 𝑋3 ) 1 0.6
Error 20 16.9 0.845
Total 24 124.8

𝑌̂ = 𝑏0 + 𝑏1 𝑋1 + 𝑏2 𝑋2 + 𝑏3 𝑋3 + 𝑏4 𝑋4
The object is to find the appropriate model to be an estimate for the population linear
regression model. In order to do analysis, consider the following:

2.1. Find all relevant extra sums of squares


▪ 𝑆𝑆𝐸(𝑋1 )
▪ 𝑆𝑆𝐸(𝑋1 , 𝑋2 )
▪ 𝑆𝑆𝐸(𝑋1 , 𝑋2 , 𝑋3 )
▪ 𝑆𝑆𝑅(𝑋1 , 𝑋2 , 𝑋3 , 𝑋4 ) and 𝑆𝑆𝐸(𝑋1 , 𝑋2 , 𝑋3 , 𝑋4 )

2.2. Is there a linear relationship between the response and a set of predictors?

1
2.3. Find the coefficient of multiple determination and explain what it measures.

2.4. Find all the coefficients of partial determination and explain each of them.

2.5. Confirm the above results by testing whether or not 𝑋3 𝑎𝑛𝑑 𝑋4 are necessary. Also, give the fitted
line that should now be used.

2.6. Explain the difference between the coefficient of multiple determination and a coefficient of
partial determination.

2.7. Let us assume that the above scenario is based on the following;

𝑋1 : Number of Beds (NoB)


𝑋2 : Income per bed occupied in R1000.00 (IpB)
𝑋3 : Salaries of staff in R10 000.00 (Sal)
𝑋4 : Number of Doctors employed (NoD)

𝑌̂: Income of hospital in R1 000 000.00

The company that owns 400 hospitals provided the data. The table contains an extract thereof.

𝑌𝑖 𝑋1 𝑋2 (R1000) 𝑋3 (𝑅10 000) 𝑋4


14.5 200 1.5 11 20
12 150 1.8 8 15
12.5 400 0.9 12 15
10.5 350 0.85 10 19

A regression model was found to fit the data and the least square method provided the
following fitted response function.

𝑌̂ = 15 + 0.12𝑋1 + 1.45𝑋2 − 1.2𝑋3 − 0.9𝑋4

2.7.1. What is the income of the hospital if 𝑋1 = 400, 𝑋2 = 0.9, 𝑋3 =12 and 𝑋4 =15?

2.7.2. Interpret the coefficient of 𝑏3 and 𝑏4 .

2.7.3. Explain the following.


2
𝑅𝑌3|12 = 0.0932

2.7.4. Rewrite the conclusion below using variable names, from the hypothesis test in 2.5. Do
you have the new model based on the conclusions in 2.7.3? Motivate your answer.

Conclusion: 𝐹 ∗ < 𝐹 and small we have concluded that 𝛽3 = 𝛽4 = 0 and thus the
predictors 𝑋3 𝑎𝑛𝑑 𝑋4 can be taken out of the regression model.

___________________________________________________________________________
The End.

You might also like