Table 7

RESULTS OF GENETIC ADAPTIVE NEURAL NETWORK (GANN) APPROXIMATION

OF EURODOLLAR PUT FUTURES OPTION TRADED ON LIFFE

Panel A: Training Data Set

   

CS OPM Approximation

GANN Approximation

Sub-sample

Description

Number of

Observations

Mean Squared

Error (MSE)

Mean Absolute

Error (MAE)

Mean Squared

Error (MSE)

Mean Absolute

Error (MAE)

Complete

Sample

2000

0.02012

0.07743

0.00044

0.01598

M* > .01

1202

0.01348

0.05568

0.00039

0.01511

.01 < M < 1

577

0.02641

0.09886

0.00047

0.01648

M > 1

619

0.00155

0.01575

0.00031

0.01377

M < -.01

790

0.02939

0.10903

0.00051

0.01726

-1 < M < -.01

517

0.03381

0.11526

0.00061

0.01872

M < -1

271

0.02116

0.09785

0.00032

0.01442

-.01 < M < .01

8

0.10399

0.22549

0.00129

0.02194

Panel B: Holdout Data Set, HOLDOUT1

Complete

Sample

6887

0.01759

0.07162

0.00046

0.01632

M* > .01

4159

0.01051

0.04892

0.00041

0.01511

.01 < M < 1

1840

0.02174

0.09017

0.00056

0.01740

M > 1

2312

0.00159

0.01608

0.00029

0.01328

M < -.01

2698

0.02822

0.10593

0.00054

0.01806

-1 < M < -.01

1782

0.03290

0.11229

0.00062

0.01898

M < -1

910

0.01912

0.09361

0.00039

0.01627

-.01 < M < .01

30

0.04373

0.13312

0.00118

0.02606

* M is a measure of the degree of moneyness. For example, if the futures rate (F(t)) is greater than the strike rate (100 - K), a put option on a 3-month Eurodollar futures contract will be in-the-money. Therefore, M is defined as F(t) - (100 - K), and when M > .01 the put option is in-the-money.

[Back to Document]