Business Forecasting

Description

Project Description

For this project, you will develop and analyze a multiple regression model to forecast a business-related outcome using real-world data. This assignment aims to enhance your skills in statistical modeling, data analysis, and critical interpretation of forecast results.

You will select a dataset comprised of one dependent variable and four (4) independent variables. You can choose any dependent variable you would like to study, as long as you can find at least 20 observations.

The four independent variables MUST BE macroeconomic variables. Forecasted values of macroeconomic variables are readily available, making it relatively easy to make forecasts of your chosen dependent variable.

Project Components and Guidelines

You will prepare a report (including tables and graphs) that summarize the findings of your regression analysis of your data:

1. Clearly articulate the forecasting question(s) you are you are attempting to address with your model.

2. Justify your choice of predictor variables based on business/economic relevance.

3. Create and present a table of summary statistics of your dataset.

4. Create a correlogram of your dependent variables.

5. Provide plots of your dependent variable against each independent variable. (You will have a sperate graph for each independent variable.)

6. Estimate the model parameters using the regression tool in Excel’s Analysis ToolPak addon.

7. Provide a business/economic interpretation of each estimated coefficient.

8. Discuss the model diagnostics (t-tests, p-values, R-squared, adjusted R-squared, F statistic, and Durbin-Watson statistic.)

9. Use your parameter estimates to generate estimates of your dependent variable for the next five (5) periods.

10. Prepare a presentation to present to the class.

Your report and presentation are due in class on Tuesday, November 26. Presentations (chosen in a random order) will also begin that day.

Data Sources

Of course, one place that you can find data is on the Federal Reserve Bank of St. Louis’s FRED database. You can search FRED using phrases such as:

1. Industry Sales Data: “sales data [industry name]”
2. Industry Revenue: “revenue [industry name]”
3. Industry Valuation Multiples: “valuation multiples [industry name]”
4. Industry Financial Ratios: “financial ratios [industry name]”
5. Industry Economic Indicators: “economic indicators [industry name]”
6. Industry Sector Analysis: “sector analysis [industry name]”
7. Or simply search for “sales” or “transactions”.

Other sources of data:

U.S. Bureau of Labor Statistics has data on employment rates, wages, and productivity measures.

World Bank Open Data has international economic indicators.

Data.gov has data on retail sales and industry-specific indicators.

If you are really motivated, take a look at the UCI Machine Learning Repository.

I have posted the Congressional Budget Office’s macroeconomic forecasts in a spreadsheet in this module. You can use this data to make forecasts based on your regression results.

I have just done the data but read all these clearly and do the presentation and three paper report