Business Forecasting using Forecast X software

Business Forecasting using Forecast X software

Research Paper Instructions

The following step by step instructionsare based on the sample paper.  For you paper, you could choose a different business/economic variable and different historical, holdout, and forecasting periods.  If you choose to use NHS, you need to use the most current data.

1. Define the chosen variable

• New one-family houses sold (NHS)

2. Collect the data for the variable

• Data source
• economagic.com(For this site, you will need to enter the data by yourself unless you pay the fee)
• https://www.stlouisfed.org/(For this site, you can download the data to the Excel free) or other sources
• Monthly data and not seasonally adjusted (NSA)
• Time period –January 1975 to December 2009

3. Enter the data in Excel according to ForecastX format

• 1/1/1975 for Jan. 1975 and 2/1/1975 for Feb. 1975

4. Plot the graphfor whole series

• ForecastX — Preview

5. ACF for the whole series

• ForecastX –Analyze
• Identify the trend

6. ACF for the first differenced series

• ForecastX – Analyze – Differencing – Non-Seasonal >1
• Identify seasonality

7. Decide the sample period for model selection

• Historical period: 1/1/1975-6/1/2009
• Holdout period: 7/1/2009-12/1/2009

8. Select models according to the data pattern

• Table 2.1, p. 58
• Time-series models: Modified naïve model, Winters’exponential smoothing, Time-series Decomposition, and ARIMA
• Regression mode:
• NHS = f (IR, DPI, dummy variables)—need data for IR (30-year mortgage rate), DPI(disposable personal income) and dummy variables (for seasonality)

9. Perform estimation and forecasting for the data in the historical period, 1/2001 – 12/2010

• ForecastX – Forecast Method – choose models
• ForecastX – Statistics — RMSE
• For regression model, standard report shows the forecasts for independent variables

10. Compare MAPEs and RMSEs of different models for the historical and holdout periods
11. Perform ex-ante forecast – 1/1/2010 – 6/1/2010

• Use the whole series, 6/1/1975 – 12/1/2009
• Either choose the best model or combine two models