1. Computer Project 3:
- Regression and Correlation Analysis
The format is similar to project#1, which starts with a professional looking cover page,
followed by the table of content, then the analysis and the appendix sections.
The analysis section should include the following sections:
1. 1- Model Building: Explain your model in terms of the “cause and effect” i.e.
clarify the dependent and the independent variables in your model. State and
explain your hypothesis (theory) about a possible relationship between your dependent and the independent
variables i.e. the type and the direction of the relationship between these variables
should be explained. Example “there is a significant direct (or indirect) linear
relationship between Var1 and Var2”. If you wanted to develop a
multi-variable model, what other factors you would have included in your
model. Explain the determination of the independent and the dependent
variable in your model.
2. 2- Visualization Step: Develop a scatter diagram and conclude if there appears to be
a linear or a non-linear relationship between your variables.
3. 3- Estimation Step: Estimate the regression line. Calculate and INTERPRET the
estimated slope Calculate and INTERPRET the estimated intercept.
4. 4- Goodness of Fit: Calculate the coefficient of determination and comment on the
goodness of fit. Interpret your finding.
5. 5- Strength of linear relationship: Calculate and INTERPRET the estimated correlation
coefficient. Comment on the measure of strength of the linear relationship.
6. 6- Test of the Strength: Test the significance of the strength of linear relationship
between the two variables at 1% and then at 10% level of significance.
Interpret your finding. Comment on the p-value.
7. 7- Test of the Significance: Test the significance of the linear relationship between the
two variables at 5%and then at 10% level of significance. Interpret your
finding. Comment on the P-value.
8. 8- Prediction Step: Use the estimated regression equation and forecast the value of
Y based on a given value of X.
9- If you wanted to develop a multi-variable model, what other factors you would have
included in your model
10- EXTRA CREDIT- Develop a multivariate model of at least 2 independent variables.
Compare to your earlier model, do you get more explanatory power? Comment on the
coefficients and check the significance of the model as well as each independent
variable.