Air Travel Forecast Problem Objectives Introduction to forecasting - - PowerPoint PPT Presentation

1

Air Travel Forecast Problem

Air Travel Forecast Problem

Objectives

• Introduction to forecasting methods
• Experience with Delphi
• Experience with consensus-seeking techniques
• Strength/weaknesses of various methods

2

Air Travel Forecast Problem

Methodology Tree for Forecasting

Causal models Data mining Statistical Univariate Theory- based

Data- based

Extrapolation models Multivariate Rule-based forecasting Unaided judgment Judgmental Self Others Role playing (Simulated interaction) Role No role Conjoint analysis Knowledge source Quantitative analogies Unstructured Structured Feedback No feedback Prediction markets Delphi Decom- position Structured analogies

Methodology Tree for Forecasting forecastingpriciples.com JSA-KCG September 2005

Neural nets Expert systems Intentions/ expectations Judgmental bootstrapping Segmentation Linear Classification Game theory

3

Air Travel Forecast Problem

Techniques for Forecasting

Form groups of about 5 to 7 people, then use the: Delphi procedure

First estimate – individual and anonymous Statistical summary – group Group discussion (use consensus technique) Second estimate – individual and anonymous Statistical summary - group

Minutes

12 3 20 2 3 40

4

Air Travel Forecast Problem

Group Results

Extrapolations Causal model Segmentation Bootstrapping Judgment Average ranks 5 4 3 2 1 Accuracy Rankings: (Round 2) Group

5

Air Travel Forecast Problem

Discussion

Discuss Delphi Expected results When to use Actual Results Initial hypotheses Results in Air Travel study Calculation of your error score Conclusions

6

Air Travel Forecast Problem

Delphi

Agreement among experts Your results More agreement among panelists on Round 1 _____ No differences (Round 1 vs. 2) _____ More agreement on Round 2 _____ Findings from literature: Typically more agreement on later rounds Expected accuracy: Which do you expect to be closest to actual ranks? Your opinions Round 1 more accurate _____ Round 2 more accurate _____ No difference _____ Delphi improves accuracy vs. traditional meetings given some expertise among panelists

7

Air Travel Forecast Problem

Round 2: Previous Rankings vs. Your Rankings

4.8 4.7 Extrapolation 2.9 2.6 Causal 2.0 2.2 Segmentation 2.9 3.2 Bootstrapping 2.4 2.2 Judgment You

• Adv. Mgmt.

(28 groups)* MBA (21 groups)* Method Average Ranking

*Groups from U.S., Sweden, Norway, and Netherlands

8

Air Travel Forecast Problem

Evidence-based Findings

(“>” means “more accurate than”)

• 1. Objective methods > subjective: especially for large

changes

• 2. Causal methods > naïve: especially for large changes
• 3. Bootstrapping > Judgment
• 4. Structured meetings > unstructured

No Yes Sufficient

• bjective data

Yes No Yes No Large changes expected Policy analysis Yes No Conflict among a few decision makers Type of knowledge Policy analysis No Yes Domain Self Yes No Time series Cross-section Type of data Good knowledge of relationships Policy analysis No Yes Good domain knowledge Yes No Yes No Large changes likely Similar cases exist Yes No Judgmental methods Quantitative methods Yes No Delphi/ Prediction markets Judgmental bootstrapping/ Decomposition Conjoint analysis Intentions/ expectations Role playing (Simulated interaction/ Game theory) Structured analogies Expert systems Rule-based forecasting Extrapolation/ Neural nets/ Data mining Causal models/ Segmentation Quantitative analogies Accuracy feedback Unaided judgment No Yes Selection Tree for Forecasting Methods forecastingprinciples.com JSA-KCG January 2006 Yes No Use adjusted forecast Several methods provide useful forecasts Single method Omitted information? Combine forecasts Use unadjusted forecast

Using the Selection Tree

?

9

10

Air Travel Forecast Problem

Rankings based on Evidence-based Findings

Subjective and causal 5 Judgment Objective/subjective and causal 4 Bootstrapping Objective and naïve 3 Extrapolation 1.5 Segmentation Objective and causal 1.5 Causal model Why? Rank Method

Evidence summarized in Armstrong (1985), Long-Range Forecasting, and Armstrong (2001), Principles of Forecasting – see forecastingprinciples.com

11

Air Travel Forecast Problem

Accuracy of the Different Methods of Forecasting U.S. Air Travel, 1963-1968

(Successive updating used)

Source: Armstrong & Grohman (1972) in full text at forecastingprinciples.com

5.8 28.1 19.3 Averages (21) 4.2 6.8 7.3 9.8 6.2 0.7 6.8 15.6 25.1 34.1 42.1 45.0** 5.7 12.7 17.4 22.5 27.5 29.9 1 (6) 2 (5) 3 (4) 4 (3) 5 (2) 6 (1) Econometric Judgment Extrapolation Mean Absolute Percentage Error* Forecast Horizon

Years (Number Ahead of Forecasts) * The forecasts were lower than actual in nearly all cases. ** Estimated

12

Air Travel Forecast Problem

Average Error Scores*

Round 2 MBAs 7.4 Advanced Mgt. 7.5 Forecasting Experts 8.4 You

*Key: Best possible = 0 No information (all ties) = 6 Worst possible = 12

13

Air Travel Forecast Problem

General Advice

• Beware of unaided judgment
• Be conservative when uncertain – thus, use equal ranks

given uncertainty about most accurate method