FIN4060 Week 2 Project

Quantitative Forecasting

Quantitative forecasts are number based. They may be simple or may rely heavily on statistical

methods. 

You can determine the quality of a forecast by calculating a number of different measures of forecast

accuracy.  Among the most widely used are the mean absolute deviation (MAD), the mean squared error (MSE), or the mean absolute percent error (MAPE).

Besides forecast accuracy, other factors also should be considered when evaluating a forecast. For

example, is the data seasonal? What are the current trends? Are the customers' preferences changing?

These factors can have an impact on the forecast and need to be included. Seasonality can be included

through the use of a seasonal index that relates the average demand in a period to the average demand

in all periods.

A forecast can be created by graphing a series of historical data points and then �tting a line to the

series to predict future values. This graph is considered a trend projection because it assumes the

future will follow the current trend and the same path.

Another technique that helps you forecast demand is regression. This technique assumes a linear

relationship between the independent and dependent variables. Regression differs from other

forecasting techniques because it can provide a distribution of possible values rather than a single

value. This distribution is referred to as the standard error of the estimate. In addition, a regression equation indicates through the coef�cient of correlation how closely the model represents your data.

Quantitative forecasts require calculation.  See the Supplemental Media entitled “Forecasts and

Errors” in order to review multiple forecasting techniques and the calculations associated with the

measures of forecast error listed in the video. 

Additional Materials

Forecasts and Errors (media/week2/SUO_MGT3059%20W2%20L3%20Forecasts%20And%20Errors.pdf?

_&d2lSessionVal=SWzqLkE3HvLXZkZ375Dqp03nU&ou=86458)

