Palisade2

Using NeuralToolsto generate

a pricing model for wool

Kimbal Curtis and John Stanton

Australian Wool Industry

70% of world trade in apparel wool is Australian woolUnlike other commodities• Each farm lot is fully measured• Each farm lot has an individual priceAbout 450,000 farm lots sold each year in AustraliaRaw wool value of AUD3 billion annually

Wool prices & market reporting

Estimates of auction price on individual lots needed by sellers (farmers)

Forecast auction price on individual lots required by buyers for contracts

Market reporting of price paid for different wool types

Neural nets & wool prices

Neural nets attractive because• Number of records is large• Prices are dynamic• Price/attribute relationships are non-linear and

interactive• Price/attribute relationships are dynamic over

time• The data set is incomplete and imprecise

All Merino fleece lots

(Fremantle Jan-Mar 2006)Each grey dot represents a parcel of wool

sold at auction i.e. a ‘case’

Long & short fleece lots

(Fremantle Jan-Mar 2006)Long and short wool

differentiated on price

Merino pieces lots

(Fremantle Jan-Mar 2006)Pieces wool

(a subset of the wool clip)

The Challenge !

(Fremantle Jan-Mar 2006)

Market Indicators

Market indicators, like a stock market index, used to price wool

Model development (1)

Assemble 6 month data set• Independent category and numeric variables• Dependent numeric variable (price)• Training, testing and prediction data

Use Best Net SearchEvaluate predictive capabilityRefine model

Excel Demo


Assemble a 6 month data setUse Best Net Search• GRN – proved best in most cases

(generalised regression neural net)• MLFN – also tried with up to 5 nodes

(multi layer feed-forward neural net)

Evaluate predictive capabilityRefine model

Configuration summary

Net Information Name Net Trained on Pieces wool sales, weeks 33 -

38, 2006 (3) Configurations Included in Search GRNN, MLFN 2 to 3 nodes Best Configuration GRNN Numeric Predictor Location Palisade Conf Curtis v6 BNS 6hrs.xls Independent Category Variables 8 (Sale centre, Sale week, Sale outcome,

Style, Med Hard Cotts, Unscourable Colour, Jowls, Dark Stain)

Independent Numeric Variables 8 (Staple Length, Staple Strength, Vegetable Matter, Diameter, CV Diameter, Mid Breaks, Yield, Hauteur)

Dependent Variable Numeric Var. (Clean price)


Assemble a 6 month data setUse Best Net SearchEvaluate predictive capabilityRefine model

Model evaluation (1)

NeuralTools outputs• Error measures• Actual versus Predicted, Residuals• Variable Impact Analysis

Live PredictionRelationships between variablesCompare to published market indicators

Model evaluation (1)Training and Testing summary

Training Number of Cases 5910 Training Time (h:min:sec) 0:39:43 Number of Trials 104 Reason Stopped Auto-Stopped % Bad Predictions (5% Tolerance) 14.7377% Root Mean Square Error 24.72 Mean Absolute Error 16.42 Std. Deviation of Abs. Error 18.48Testing Number of Cases 1507 % Bad Predictions (5% Tolerance) 43.3975% Root Mean Square Error 53.18 Mean Absolute Error 36.99 Std. Deviation of Abs. Error 38.21

Model evaluation - Training data(mean absolute error 16 cents)

Model evaluation - Testing data(mean absolute error 37 cents)

Model evaluation (1)Testing data (indicators)

Observed versus predicted for the published Pieces Market indicators

Most points are on the 1:1 line, but a small group hover above i.e. they have higher predicted values than reported

Model evaluation (1)Variable impact analysis

Relative Variable Impacts

41.3%18.7%

11.7%8.8%

7.7%1.9%1.8%1.6%1.2%1.2%1.1%0.9%0.7%0.6%0.4%0.4%

0% 10% 20% 30% 40% 50% 60% 70%

Diameter Vegetable Matter

Staple Length Jowls

Hauteur Sale outcome

Med Hard Cotts Yield

CV Diameter Staple Strength

Sale centre Sale week Dark Stain

Style Unscourable Colour

Mid Breaks

This is a sensitivity analysis, not the percent of varianceaccounted for by each variable




Model evaluation (2)Live prediction

Sale centre FremantleSale week W38

Style Average

Med Hard Cotts C0Unscourable Colour H0Jowls J0Dark Stain S0

Diameter 20.0

Yield 50.0Vegetable Matter 2.5

Staple Length 80Staple Strength 35Mid Breaks 55Hauteur 62

Clean price 664

Simple spreadsheet pricing tool.

Change any of the values in the yellow cells, and ‘Live prediction’updates the clean price




Model evaluation (3)relationships between variables

6570

7580

85

25

30

3540

45

590

600

610

620

630

640

650

660

670

680

CleanPrice

StapleLength

StapleStrength

SydneyWeek 3821 micron2% VM


Fremantle Melbourne Sydney

6570

7580

85

25

3035

4045

595

600

605

610

615

620

625

630

635

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

500

520

540

560

580

600

620

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

600

610

620

630

640

650

660

670

680

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

590

600

610

620

630

640

650

660

670

680

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

560

570

580

590

600

610

620

630

640

650

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

540

550

560

570

580

590

600

610

620

630

640

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

21 m

icro

n22

mic

ron


Fremantle Melbourne Sydney

6570

7580

85

25

3035

4045

705

710

715

720

725

730

735

740

745

750

755

760

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

750

760

770

780

790

800

810

820

830

840

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

740

750

760

770

780

790

800

810

820

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

640

645

650

655

660

665

670

675

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

670

680

690

700

710

720

730

740

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

6570

7580

85

25

3035

4045

660

670

680

690

700

710

720

Cl eanP r i ce

St apl eLengt h

St apl eSt r engt h

19 m

icro

n20

mic

ron

Price spread variation




Model evaluation (4)predictive capability

20 micron indicator

22 micron indicatorMelbourneWeek 38


Dark blue lots have SL, SS and VM “similar” to market indicator definition

MelbourneWeek 38


MelbourneWeek 37


MelbourneWeek 37


MelbourneWeek 36


MelbourneWeek 35


MelbourneWeek 34


MelbourneWeek 33


FremantleWeek 37


FremantleWeek 38


Assemble a 6 month data setUse Best Net SearchEvaluate predictive capabilityRefine model • Reduce variables• Combine selling centres• Sale week - category variable

Some Neural Net applications

Market reportingPrice predictorValidation check for other estimatesMissing sale problemGenerate price matrices• Using Live Prediction and @Risk

Summary

Data rich application with characteristics that looked ideal for NeuralToolsSolutions generated which can support industry analysis and generation of indicators

Documents

Palisade2