20
ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Embed Size (px)

Citation preview

Page 1: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD

Prepared by Dan Milewski

November 29, 2005

Page 2: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Tutorial Outline

1. Defining the Method

2. When to Use the Method

3. How to Use the Method

4. An Example

5. An Exercise

6. Summary

7. Readings List

Page 3: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Defining the Method

A Forecasting Model:

• Predicts future levels of a variable

• Can be either quantitative or qualitative

1 2 3 4 5 6 7

Page 4: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Defining the Method

Exponential Smoothing:

• Quantitative forecasting method

• Weighted average of two variables

1 2 3 4 5 6 7

Page 5: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Defining the Method

Adjusted

• Trend adjustment factor included

• Better at picking up on trends

1 2 3 4 5 6 7

Page 6: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Defining the Method

So, combined,….

Adjusted Exponential Smoothing Forecasting Method:

A method that uses measurable, historical data observations, to make forecasts by calculating the weighted average of the current period’s actual value and forecast, with a trend adjustment added in.

1 2 3 4 5 6 7

Page 7: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

When to Use the Method

• Preferred Scenario:– When a trend is present

• Good Scenario:– When there’s a cyclical or seasonal pattern

• Least-effective Scenario– Working with random variations

1 2 3 4 5 6 7

Page 8: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

When to Use the Method

1 2 3 4 5 6 7

Page 9: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

When to Use the Method

• Manufacturing Firms:– To forecast demand

• Service Organizations:– To forecast customer arrival patterns

• Financial Analysts:– To forecast revenues and profits

• Investors:– To forecast economic indicators

1 2 3 4 5 6 7

Page 10: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

How to Use the Method

Exponential Smoothing:Exponential Smoothing:

FFtt+1+1 = = DDtt + (1 - ) + (1 - )FFtt

Where…Where…

FFt t +1 +1 = = forecast for next periodforecast for next period

DDtt = = actual value for present periodactual value for present period

FFtt = = previously determined forecast for previously determined forecast for

present periodpresent period

== weighting factor (between 0 and 1)weighting factor (between 0 and 1)

1 2 3 4 5 6 7

Page 11: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

How to Use the Method

Adjusted Exponential Smoothing:Adjusted Exponential Smoothing:

AFAFtt+1+1 = = FFtt+1+1 + T + Ttt+1+1

Where…Where… TTt t +1 +1 = ( = (FFtt+1+1 – – FFt t )) + (1 - ) + (1 - ) TTtt

= trend factor for the next period= trend factor for the next period TTtt = trend factor for the current period = trend factor for the current period = smoothing constant for the trend = smoothing constant for the trend

adjustment factoradjustment factor

(just add a trend adjustment factor)(just add a trend adjustment factor)1 2 3 4 5 6 7

Page 12: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

How to Use the Method

Points to Consider:Points to Consider:

• To start, pick an unadjusted forecastTo start, pick an unadjusted forecast

• In period 1, trend equals 0In period 1, trend equals 0

1 2 3 4 5 6 7

Page 13: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

An Example

2005 U.S. Housing Starts (monthly):2005 U.S. Housing Starts (monthly):

1 2 3 4 5 6 7

Page 14: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

An Example

2005 U.S. Housing Starts (monthly):2005 U.S. Housing Starts (monthly):

1 2 3 4 5 6 7

Page 15: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

An Exercise

Using the adjusted exponential smoothing forecasting method and the following data…

– Predict Q4 2005 sales revenues for Intel • Where = 0.4 and = 0.7

– Predict Q4 2005 net income for Intel• Where = 0.2 and = 0.6

1 2 3 4 5 6 7

Page 16: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

An Exercise

Intel Quarterly Sales Revenue

1 2 3 4 5 6 7

Page 17: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

An Exercise

Intel Quarterly Net Income

1 2 3 4 5 6 7

Page 18: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

An Exercise

• Which series of data best fits with this method?

• What makes this so?

• What other financial data could be predicted accurately with this method?

1 2 3 4 5 6 7

Page 19: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Summary

Adjusted Exponential Smoothing Forecasting Method:

• Quantitative forecasting model

• Highly accurate

• Best when trends exist

1 2 3 4 5 6 7

Page 20: ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD Prepared by Dan Milewski November 29, 2005

Readings List

• Gardner, Jr., E.S. Exponential Smoothing: The State of the Art. Journal of Forecasting. April 1985, Vol. 3, Iss. 1.

• Jain, Chaman L. Business Forecasting Practices in 2003. The Journal of Business Forecasting Methods & Systems. Fall 2004, Vol. 23, Iss. 3

• http://home.ubalt.edu/ntsbarsh/ECON/lecture6.doc

1 2 3 4 5 6 7