Chapter 6 - Forecasting

Quantitative Methods
ADMS 3330 3 0
3330.3.0

Forecasting

QMB Chapter 6
© M.Rochon 2013

Quantitative Approaches to Forecasting
Are based on analysis of historical data concerning one or more time series.
Time series - a set of observations measured at successive points in time, or over successive periods of time.
If the historical data:
•

are restricted to past values of the series we are trying to forecast, it is a time series method.

1

Components of a Time Series
1) Trend component - the gradual shifting of the time series over a long period of time.
2) C li l component - any regular sequential
Cyclical
t l ti l pattern of values above and below the trend line.
3) Seasonal component - regular patterns of variability within certain time periods, such as over a year.
4) Irregular component - short-term, unanticipated and non-recurring factors that affect the values of the time series. One cannot predict irregular component impact on the time series.

Time Series Methods
The three time series forecasting methods are: 1) Smoothing Methods
2) Trend Projection
3) Trend Projection, adjusted for seasonal influence (Multiplicative Model)

2

Smoothing Methods
Smoothing methods are used to average out the irregular components of the time series in cases where the time series: is fairly stable, and has no significant trend, seasonal, or cyclical effects. •
•

Four common smoothing methods:
1)
2)
3)
4)

Moving Average
Weighted Moving Averages
Exponential Smoothing
Centered Moving Average (not for forecasting as

we will see later – only a process to lead to forecasting)

Measures of Forecast Accuracy
Mean Squared Error (MSE)
•
•
•

Find the forecast errors. Square them.
Calculate the mean mean. Select forecasting method which minimizes MSE.

3

Example: Robert’s Drugs
During the past ten weeks, sales of cases of
Comfort headache medicine at Robert's Drugs have been:
Week Sales