Question: Forecasting Models

The Consolidated Edison Company sells electricity (82% of revenues), gas (13%) and steam (5%) in New York City and Westchester County. Suppose now is in early January of year 2002, you are working in the Finance Dept. of the company, and the CFO needs to conduct a budget plan for the next eight quarters (Quarter 1 of 2002 to Quarter 4 of 2003). You are required to provide valid estimates of total revenues for these eight quarters. The historical revenue records for last 68 quarters (1985-Q1 to 2001-Q4) are provided in the file of Consolidated_Edison_Co.xls.

1. . Plot the time series data of total revenues and identify any patterns you find.
2. . Based on the patterns you found, please construct an appropriate exponential smoothing model to capture the behaviors of the time series. If you use Holt’s model, please use initial level and trend estimates of: L ₀ = 1441 and T ₀ = 0. If you use Winters’ model, besides the same initial level and trend estimates
   (L ₀ = 1441 and T ₀ = 0), initial seasonal factors are set to be 1.0: S _-3 = 1.0, S _-2 = 1.0, S _-1 = 1.0 and S ₀ = 1.0. Note: Please use = = = 0.50 as parameters’ values for constructing either Holt’s or Winters’ model.
3. . Construct an appropriate multiple regression model for the total revenues series.
4. . Between the two models constructed above, which one is better for prediction purposes? Why?
5. . Based on the better model you select in (4), please list your predictions of total revenues for the next eight quarters.

Price: $2.99

Solution: The downloadable solution consists of 9 pages
Deliverable: Word Document