Adjusted Exponential Smoothing 5c60d

ADJUSTED EXPONENTIAL SMOOTHING FORECASTING METHOD

Prepared by Dan Milewski November 29, 2005

Tutorial Outline 1. 2. 3. 4. 5. 6. 7.

Defining the Method When to Use the Method How to Use the Method An Example An Exercise Summary Readings List

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

Defining the Method Exponential Smoothing:

• Quantitative forecasting method • Weighted average of two variables

1

2

3

4

5

6

7

Defining the Method Adjusted

• Trend adjustment factor included • Better at picking up on trends

1

2

3

4

5

6

7

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

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

When to Use the Method

1

2

3

4

5

6

7

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

How to Use the Method Exponential Smoothing: Ft+1 = Where… Ft +1 = Dt = Ft =

Dt + (1 -

)Ft

forecast for next period actual value for present period previously determined forecast for present period = weighting factor (between 0 and 1) 1

2

3

4

5

6

7

How to Use the Method Adjusted Exponential Smoothing: AFt+1 = Ft+1 + Tt+1 Where… Tt +1 = (Ft+1 – Ft ) + (1 - ) Tt = trend factor for the next period Tt = trend factor for the current period = smoothing constant for the trend adjustment factor (just add a trend adjustment factor) 1

2

3

4

5

6

7

How to Use the Method Points to Consider: • To start, pick an unadjusted forecast • In period 1, trend equals 0

1

2

3

4

5

6

7

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

1

2

3

4

5

6

7

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

1

2

3

4

5

6

7

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

An Exercise Intel Quarterly Sales Revenue

1

2

3

4

5

6

7

An Exercise Intel Quarterly Net Income

1

2

3

4

5

6

7

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

Summary Adjusted Exponential Smoothing Forecasting Method:

• Quantitative forecasting model • Highly accurate

• Best when trends exist

1

2

3

4

5

6

7

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