Description

(100 points) Using the data supplied from ECONOMAGIC (see Excel attachment), estimate a forecasting model for the Number of New Homes Sold in the U.S (NHS) measured in thousands of units as the dependent variable (Y). As your independent variables (X’s) use the Real Disposable Income ($B) and the Unemployment Rate (%).  The data are collected Monthly beginning in January 1963 and ending August 2019.

1.      Initially, graph a time series for each variable and give a brief description of each variable's history.

2.      Graph a 3-dimensional function using the NHS as the ‘z’ variable and the Real Disposable Income and Unemployment Rate as the ‘x’ and ‘y’ variables.

3.      Test for stationarity in the dependent (NHS) variable only. Briefly discuss.

4.      If data are non-stationary correct the model and briefly explain.

5.      Next, estimate and interpret a multiple regression equation. In your answer make sure to include a discussion of the regression coefficients and all supporting statistics (i.e. t-tests, R-square, F-test, etc.).

6.      Closely inspect the residuals from the regression equation.  Are there any discernible patterns in these residuals with particular reference to outliers and/or leverages? Briefly explain

7.      Now using this estimated regression model, provide and briefly interpret for the reader three ‘point’ and ‘interval’ predictions or forecasts (i.e., for the NHS).  Use the following specified values for the independent variables (selected to represent a steady state, a pessimistic, and an optimistic economic environment).

a. Real Disposable Income =$15,000B   UnempRate=3.70% (steady state forecast)

b. Real Disposable Income =$14,000B   UnempRate=5.0% (pessimistic forecast)

c. Real Disposable Income =$16,000B   UnempRate=2.5% (optimistic forecast)

8.      Now re-estimate the model using a double logarithmic functional form (columns G-I) are the appropriate transformations in your Excel worksheet) and interpret your regression equation (i.e. regression coefficients, t-tests, R-square, F-test, residuals outliers, leverages, etc.).

9.      Compare and contrast the results from the logarithmic model (#8) with the unlogged model (#5).

10.  Test to determine if the estimated slope coefficient on the ln(Income) variable is equal to ‘1.00’; that is, is there a one-to-one correspondence between NHS and Real Disposable Income?

Submit your paper to this assignment link as an attached WORD document and clearly label your answers as  ordered above  (i.e., 1-10).  Remember you may copy and paste the relevant MINITAB/Economagic outputs but your final analysis must be in essay format!

For your review, I have attached an EXCEL and MINITAB documents created from ECONOMAGIC that displays all the relevant data (including the raw as well as the logarithmic transformation).  Use the MINITAB document to complete the test

Also please insert your name at the top of the WORD document and make sure when you cut and paste your relevant MINITAB output that the resulting output is properly aligned.

I will start a THREAD on the Discussion Board if you have questions concerning the Midterm