1. Develop the following six regression models to predict the number of housing starts:
2. Linear regression model using the number of housing starts as the dependent variable and mortgage rate as the independent variable. Label your results in an Excel workbook using the prompt number.Write the regression equation for the model.
3. Quadratic regression model using the number of housing starts as the dependent variable and mortgage rate and mortgage rate² as independent variables. Label your results in an Excel workbook using the prompt number.Write the regression equation for the model.
4. Cubic regression model using the number of housing starts as the dependent variable and mortgage rate, mortgage rate², and mortgage rate³ as independent variables. Label your results in an Excel workbook using the prompt number.Write the regression equation for the model.
5. Log-log regression model using the natural logarithm of the number of housing starts as the dependent variable and the natural logarithm of mortgage rate as the independent variable. Label your results in an Excel workbook using the prompt number.Write the regression equation for the model.
6. Logarithmic regression model using the number of housing starts as the dependent variable and the natural logarithm of mortgage rate as the independent variable. Label your results in an Excel workbook using the prompt number.Write the regression equation for the model.
7. Exponential regression model using the natural logarithm of the number of housing starts as the dependent variable and mortgage rate as the independent variable. Label your results in an Excel workbook using the prompt number.Write the regression equation for the model.
8. Display the actual and predicted values for each of the above models in a time-series line graph like demonstrated in the video.You should have a separate graph for each model. Label your results in an Excel workbook using the prompt number.
9. Discuss the joint statistical significance of each of the preceding regression models at a 95% level of confidence, including identifying the appropriate regression statistics, statistic values, and criteria used to evaluate the joint statistical significance.
10. Discuss the individual statistical significance of each of the preceding regression models at a 95% level of confidence, including identifying the appropriate regression statistics, statistic values, and criteria used to evaluate the individual statistical significance.
11. Which model shows unusual findings in joint and individual significance? Describe the unusual findings. What is the likely cause of this unusual finding?
12. Check all models for the equal variance assumption. Which models that show evidence of violation of this assumption? Describe the pattern that is seen for each model that violates the assumption.
13. Compare the preceding regression models that were determined to be statistically significant and select a preferred model.Discuss the regression statistics used to compare the models, statistic values involved, and criteria used to select the preferred model.
14. Discuss the correct interpretation of the predicted change in y associated with a one unit change in x for the log-log model.
15. Using the log-log regression model, predict the number of housing starts when the mortgage rate is 5.2%.
16. Using the quadratic regression model, calculate the mortgage rate with the minimum predicted number of housing starts. Calculate the minimum predicted the number of housing starts from the model using that mortgage rate.

Date	No. Housing Starts	Mortgage Rate
2006-01-01	2273	6.15
2006-02-01	2119	6.25
2006-03-01	1969	6.32
2006-04-01	1821	6.51
2006-05-01	1942	6.60
2006-06-01	1802	6.68
2006-07-01	1737	6.76
2006-08-01	1650	6.52
2006-09-01	1720	6.40
2006-10-01	1491	6.36
2006-11-01	1570	6.24
2006-12-01	1649	6.14
2007-01-01	1409	6.22
2007-02-01	1480	6.29
2007-03-01	1495	6.16
2007-04-01	1490	6.18
2007-05-01	1415	6.26
2007-06-01	1448	6.66
2007-07-01	1354	6.70
2007-08-01	1330	6.57
2007-09-01	1183	6.38
2007-10-01	1264	6.38
2007-11-01	1197	6.21
2007-12-01	1037	6.10
2008-01-01	1084	5.76
2008-02-01	1103	5.92
2008-03-01	1005	5.97
2008-04-01	1013	5.92
2008-05-01	973	6.04
2008-06-01	1046	6.32
2008-07-01	923	6.43
2008-08-01	844	6.48
2008-09-01	820	6.04
2008-10-01	777	6.20
2008-11-01	652	6.09
2008-12-01	560	5.33
2009-01-01	488	5.06
2009-02-01	581	5.13
2009-03-01	520	5.00
2009-04-01	477	4.81
2009-05-01	550	4.86
2009-06-01	583	5.42
2009-07-01	587	5.22
2009-08-01	585	5.19
2009-09-01	586	5.06
2009-10-01	529	4.95
2009-11-01	589	4.88
2009-12-01	576	4.93
2010-01-01	612	5.03
2010-02-01	605	4.99
2010-03-01	634	4.97
2010-04-01	679	5.10
2010-05-01	588	4.89
2010-06-01	539	4.74
2010-07-01	550	4.56
2010-08-01	614	4.43
2010-09-01	601	4.35
2010-10-01	533	4.23
2010-11-01	548	4.30
2010-12-01	520	4.71