Hedging with Forwards and Futures



Similar documents
Duration and Convexity ( ) 20 = Bond B has a maturity of 5 years and also has a required rate of return of 10%. Its price is $613.

Morningstar Investor Return

Chapter 8: Regression with Lagged Explanatory Variables

PROFIT TEST MODELLING IN LIFE ASSURANCE USING SPREADSHEETS PART ONE

Principal components of stock market dynamics. Methodology and applications in brief (to be updated ) Andrei Bouzaev, bouzaev@ya.

Markit Excess Return Credit Indices Guide for price based indices

WHAT ARE OPTION CONTRACTS?

Estimating Time-Varying Equity Risk Premium The Japanese Stock Market

I. Basic Concepts (Ch. 1-4)

Chapter 7. Response of First-Order RL and RC Circuits

Risk Modelling of Collateralised Lending

INTEREST RATE FUTURES AND THEIR OPTIONS: SOME PRICING APPROACHES

Why Did the Demand for Cash Decrease Recently in Korea?

Credit Index Options: the no-armageddon pricing measure and the role of correlation after the subprime crisis

4. International Parity Conditions

Individual Health Insurance April 30, 2008 Pages

Stock Trading with Recurrent Reinforcement Learning (RRL) CS229 Application Project Gabriel Molina, SUID

Term Structure of Prices of Asian Options

SHB Gas Oil. Index Rules v1.3 Version as of 1 January 2013

Equities: Positions and Portfolio Returns

Journal Of Business & Economics Research September 2005 Volume 3, Number 9

BALANCE OF PAYMENTS. First quarter Balance of payments

Nikkei Stock Average Volatility Index Real-time Version Index Guidebook

Vector Autoregressions (VARs): Operational Perspectives

The Time Value of Money

FORWARD AND FUTURES CONTRACTS

Conceptually calculating what a 110 OTM call option should be worth if the present price of the stock is

LEASING VERSUSBUYING

Option Put-Call Parity Relations When the Underlying Security Pays Dividends

NASDAQ-100 Futures Index SM Methodology

Table of contents Chapter 1 Interest rates and factors Chapter 2 Level annuities Chapter 3 Varying annuities

Inductance and Transient Circuits

Economics Honors Exam 2008 Solutions Question 5


A Note on Using the Svensson procedure to estimate the risk free rate in corporate valuation

DELTA-GAMMA-THETA HEDGING OF CRUDE OIL ASIAN OPTIONS

11/6/2013. Chapter 14: Dynamic AD-AS. Introduction. Introduction. Keeping track of time. The model s elements

2.5 Life tables, force of mortality and standard life insurance products

Does Option Trading Have a Pervasive Impact on Underlying Stock Prices? *

Description of the CBOE S&P 500 BuyWrite Index (BXM SM )

Supplementary Appendix for Depression Babies: Do Macroeconomic Experiences Affect Risk-Taking?

INTRODUCTION TO FORECASTING

Forecasting, Ordering and Stock- Holding for Erratic Demand

The Greek financial crisis: growing imbalances and sovereign spreads. Heather D. Gibson, Stephan G. Hall and George S. Tavlas

Modeling VXX. First Version: June 2014 This Version: 13 September 2014

The yield curve, and spot and forward interest rates Moorad Choudhry

BALANCE OF PAYMENTS AND FINANCIAL MA REPORT All officiell statistik finns på: Statistikservice: tfn

Present Value Methodology

Default Risk in Equity Returns

Impact of scripless trading on business practices of Sub-brokers.

The Transport Equation

Usefulness of the Forward Curve in Forecasting Oil Prices

Distributing Human Resources among Software Development Projects 1

Chapter 9 Bond Prices and Yield

Stochastic Volatility Models: Considerations for the Lay Actuary 1. Abstract

Do Futures Lead Price Discovery in Electronic Foreign Exchange Markets?

The naive method discussed in Lecture 1 uses the most recent observations to forecast future values. That is, Y ˆ t + 1

GUIDE GOVERNING SMI RISK CONTROL INDICES

How To Calculate Price Elasiciy Per Capia Per Capi

The Interest Rate Risk of Mortgage Loan Portfolio of Banks

SURVEYING THE RELATIONSHIP BETWEEN STOCK MARKET MAKER AND LIQUIDITY IN TEHRAN STOCK EXCHANGE COMPANIES

Lecture Note on the Real Exchange Rate

Foreign Exchange and Quantos

Chapter 1.6 Financial Management

New facts in finance. John H. Cochrane

Making Use of Gate Charge Information in MOSFET and IGBT Data Sheets

Chapter 2 Problems. 3600s = 25m / s d = s t = 25m / s 0.5s = 12.5m. Δx = x(4) x(0) =12m 0m =12m

How To Price An Opion

CLASSIFICATION OF REINSURANCE IN LIFE INSURANCE

Chapter 6 Interest Rates and Bond Valuation

Small and Large Trades Around Earnings Announcements: Does Trading Behavior Explain Post-Earnings-Announcement Drift?

DYNAMIC MODELS FOR VALUATION OF WRONGFUL DEATH PAYMENTS

The Grantor Retained Annuity Trust (GRAT)

S&P GSCI Crude Oil Covered Call Index Methodology

Options and Volatility

A Tale of Two Indices

DO FUNDS FOLLOW POST-EARNINGS ANNOUNCEMENT DRIFT? RACT. Abstract

ABSTRACT KEYWORDS. Term structure, duration, uncertain cash flow, variable rates of return JEL codes: C33, E43 1. INTRODUCTION

Cointegration: The Engle and Granger approach

Chapter 4: Exponential and Logarithmic Functions

SPEC model selection algorithm for ARCH models: an options pricing evaluation framework

Chapter 6: Business Valuation (Income Approach)

1 HALF-LIFE EQUATIONS

1. y 5y + 6y = 2e t Solution: Characteristic equation is r 2 5r +6 = 0, therefore r 1 = 2, r 2 = 3, and y 1 (t) = e 2t,

CHARGE AND DISCHARGE OF A CAPACITOR

Acceleration Lab Teacher s Guide

Forecasting Sales: A Model and Some Evidence from the Retail Industry. Russell Lundholm Sarah McVay Taylor Randall

Making a Faster Cryptanalytic Time-Memory Trade-Off

THE FIRM'S INVESTMENT DECISION UNDER CERTAINTY: CAPITAL BUDGETING AND RANKING OF NEW INVESTMENT PROJECTS

Market Liquidity and the Impacts of the Computerized Trading System: Evidence from the Stock Exchange of Thailand

Relationships between Stock Prices and Accounting Information: A Review of the Residual Income and Ohlson Models. Scott Pirie* and Malcolm Smith**

Modeling VIX Futures and Pricing VIX Options in the Jump Diusion Modeling

Answer, Key Homework 2 David McIntyre Mar 25,

Transcription:

Hedging wih orwards and uures Hedging in mos cases is sraighforward. You plan o buy 10,000 barrels of oil in six monhs and you wish o eliminae he price risk. If you ake he buy-side of a forward/fuures conrac for 10,000 barrels of oil wih a mauriy of six monhs, you can eliminae he price risk. Alernaively, you are a U dollar based firm and you have a conrac from which you will receive 500,000 yen in four monhs. You plan o sell yen and buy dollars. The exchange rae risk can be eliminae by aking he sell-side of fuures conrac wih a mauriy of four monh o exchange 500,000 yen for U$ dollars a he fuures/forward rae. In each of hese examples, he price or exchange rae risk is eliminaed wih he use of a fuures conrac. The six-monh oil fuures conrac will lock in he price of he oil and he four monh yen/u$ fuures conrac will lock in he exchange rae. A mauriy he mos likely scenario will be ha in neiher case will anyone acually ake delivery of he underlying asse. or example, in he case of he oil, a mauriy he oil hedger will buy he oil on he spo marke a - T, and close ou he fuures posiion realizing a payoff of +( T T ). The resul of he hedge is a cos of - T + ( T T ) = - T. If T > T, he posiive inflow from he fuures posiion will offse par of he cos. If T < T, hen he hedger will have o pay he difference and again he ne cos of purchasing he oil will be T. In he above examples, he hedging was one for one and he mauriy of he fuures conrac exacly mached he iming of he ransacion. Ofen imes he hedging approach is no as clear as i is in hese examples. or example, he iming of he mauriy of he available fuures conracs may no be he same as he iming of he obligaion. uppose for he 10,000 barrels of oil he only fuures conrac available was for a mauriy of eigh monhs, T. If we use his o hedge our six monhs obligaion,, in six monhs we buy he oil a he spo, -, and offse he original fuures posiion by aking he sell side of he same conrac which will yield +( T ). The fuures price a ime,, will be, = e ( c y ) ( T ), This noe was prepared by Professor Rober M. Conroy. Copyrigh 003 by he Universiy of Virginia Darden chool oundaion, Charloesville, VA. All righs reserved. To order copies, send an e-mail o dardencases@virginia.edu. No par of his publicaion may be reproduced, sored in a rerieval sysem, used in a spreadshee, or ransmied in any form or by any means elecronic, mechanical, phoocopying, recording, or oherwise wihou he permission of he Darden chool oundaion.

where c is cos of carry and y is convenience yield 1 a ime. As such, he ne resul of buying he oil a spo and hedge a ime is + ( ) = T T + ( e ( c y ) ( T ) 1) T. The resul is ha he hedge is no perfec. I will depend on wha is he cos of carry and convenience yield a ime and he resul would be cerainly differen from he fixed cos of T ha we had when he mauriy of he fuures conrac exacly mached he obligaion. Hence, his mismach in mauriies creaes no quie he prefec hedge. The resuling difference from having an exac mach of mauriies is referred o a basis risk. A poenially more significan basis risk comes from a siuaion where an invesor mus use fuures conracs on a differen asse o hedge anoher asse. or example, airlines ofen wish o hedge heir je fuel coss. They sell ickes well in advance bu he acual cos of delivering he fligh will depend largely on he cos of je fuel on he dae of he ravel. Airlines can eliminae his risk by using fuures. However, hey face a problem in hedging je fuel. There are no fuures conracs raded on je fuel. The neares subsiue is heaing fuel oil. Thus, an airline could aemp o hedge heir fuel cos exposure using Heaing Oil fuures conracs. However, hey do face some risk ha he changes in he Heaing Oil fuures conracs will no exacly mach he changes in he price of Je uel. The difference beween he price of Je uel and he price of heaing Oil fuures a he dae ha he je fuel is purchased is also referred o as basis risk. As an example, Exhibi 1 shows he spo prices for je fuel and for heaing (fuel) oil from 1985 o 001. The price movemens are similar bu no quie he same. Heaing oil prices are lower and appear o be less volaile. Exhibi 1. Je uel vs. Heaing (uel) Oil 600 500 400 Price per on 300 00 100 0 Apr-85 Apr-86 Apr-87 Apr-88 Apr-89 Apr-90 Apr-91 Apr-9 Apr-93 Apr-94 Apr-95 Apr-96 Apr-97 Apr-98 Apr-99 Apr-00 Apr-01 dae Je uel ($/on -,40 lbs.) uel Oil ($/meric onne-,05lbs.) 1 Please see orward and uures noe page 7. Also Hull (5 h ediion), Chaper 3 page 60. Page

Hence using heaing oil fuures conracs on a one-o-one basis may no provide a good hedge for je fuel. or an ideal hedge, over our ime horizon we would like he change in he fuures price o exacly mach he change in he value of he asse we wish o hedge, i.e., po = uures Exhibi shows he spo prices for je fuel and for hea oil 90-day fuures and 60-day fuures. Le s assume ha an airline wishes o hedge je fuel 30 days forward in ime and he only conracs available are 90-day fuures conracs for heaing oil. The change in he spo price for je fuel over a monh is jus he price a he end of he monh less he price a he beginning of he monh. The change in he value of a fuures conrac is slighly differen. A 90-day conrac a he beginning of he monh is a 60-day conrac a he end of he monh. Hence if we use a 90-day conrac o hedge for 30 days he change in he price is he difference beween he fuures price for a 60 day conrac a he end of he monh less he fuures price for 90 day conrac a he beginning of he monh. rom exhibi, i is clear ha he price changes of he spo je fuel prices and heaing oil fuures are no he same. This raises he quesion of wheher we can use a hedge raio, h, differen from 1.0 o hedge he je fuel prices or po = h uures. Bu how do we choose he bes h? The usual soluion is o choose h such ha i minimizes he following: Min h E [( h ) ]. This resuls in a value of h ha minimizes he squared differences beween he price changes. Anoher way of saing he same hing is o choose h such ha i minimizes he variance of he hedge. In choosing h, i places a big penaly on big differences beween The minimizaion can be rewrien as E [( h ) ] = E[ ] + h E[ ] h E[ ] Assuming E[ ]=0, and E[ ]=0, hen Cov = E = E [ ], [ ], (, ) = E[ ] = ρ, ubsiuing back in he original problem resuls yields [( h ) ] = + h h ρ E, Page 3

and. Noe ha we could have chosen a very differen objecive funcion. However, his paricular objecive funcion happens o be very convenien in a number of ways. The acual soluion 3, ĥ, o his formulaion is fairly sraighforward. ˆ =, s h ρ, where is he sandard deviaion of he spo price changes, is he sandard deviaion of he fuures price changes and ρ, is he correlaion beween he spo price changes and he fuures price changes. Exhibi 3 shows he calculaion of he opimal hedge using he hisorical daa in Exhibi. The basic saisics 4 are esimaed as follows: Means: = and = = 1 1 = = 1 andard Deviaions: ( ) = and ( ) = 1 Covariance: Cov(, ) = ( ) ( ) 1 = 1 1 = 1 Correlaion: ρ, Cov = (, ) 3 The soluion o he minimizaion problem is o ake he firs derivaive of he hedge variance wih respec o h, se i equal o zero, and solve for h. E( h ) h = 0 ( + h h ρ h hˆ = ρ, ρ, = 0, ) h = 0 4 The saisics shown below are based on he populaion. If everyhing was recalculaed on a sample basis he esimaed hedge raio would be he same. Be careful using saisical funcions in excel. You need o make sure ha he esimaes of sandard deviaions and correlaions have he same basis, populaion or sample. Page 4

Hedge: ˆ s h = ρ, I is also possible o esimae he opimal hedge using regression analysis. The basic equaion is = α + h ince he basic OL regression for his equaion esimaes he value of ĥ as ˆ =, s h ρ, we can use OL regression. This is he soluion o he minimizing he original objecive funcion. Hence, his is one of he reasons ha he objecive funcion of minimizing he squared differences is so appealing. Exhibi 4 shows he oupu of an Excel regression using he daa in Exhibi 3. Noe ha he resuls are he same. The opimal hedge raio 5 is 1.064. This is very close o a value of 1.00, which is wha we would expec for wo very similar commodiies where he prices would end o move ogeher. I is useful o noe ha he regression analysis also provides us wih some informaion as o how good a hedge we are creaing. The r-square 6 of he regression ells how much of he variance in he change in spo price is explained by he variance in he change of he fuures price. In his case he r-squared saisic is.443 or 44.3%. A good hedge migh resul in an r-square value of.80. Hence, in his case, while he opimal hedge raio is close o 1.00, he hedge iself migh no be ha effecive. There is he poenial here for a lo of basis risk. Noneheless, he appropriae hedge is 1.064 heaing oil fuures conracs for each on 7 of je fuel. I have one commen on he analysis presened in his secion. Here we used he price changes in he fuures conrac for Heaing Oil. Acually, for mos pracical purposes we could have used simply he changes in he spo prices of Heaing Oil o calculae he opimal hedge. I is ofen very difficul o ge a good consisen hisoric series of fuures prices. Equiy Porfolio Hedging Hedging porfolios is he same as hedging commodiies. Consider a porfolio wih a value oday of $5,345,456. We wish o hedge his porfolio using &P 500 fuures 5 I used he excel regression funcion wih as he y variable and as he x variable. 6 The r-square of he regression is esimaed as he square of he correlaion coefficien beween and. rom exhibi 3, he correlaion coefficien is.666. quaring his yields.443. 7 Noe ha since we calculaed he opimal hedge raio based on price changes, he difference in he onnage beween he long on (,40 lbs.) for je fuel and he meric onne for heaing oil was accouned for in he analysis. Page 5

conracs. While our porfolio is similar o he &P 500, i is no he same. If we follow wha we did above, he opimal hedge is h ˆ P = ρ p, & P. & P or equiy porfolios he opimal hedge is in erms of reurns. or example, assume we have a porfolio wih a curren value of $10, 968,000. You wish o use &P 500 fuures conracs o hedge he risk over he nex monh. Exhibi 5 shows he monhly values for he porfolio and he index for he las four years. In his case, insead of using price changes we will calculae he opimal hedge raio using monhly reurns 8. rom exhibi 5, he opimal hedge raio is 1.058. Exhibi 6 shows he esimae of he hedge raio using regression analysis. Noe ha he regression model is R = α + β. P R & P This regression model is also a way o esimae Bea for a porfolio using he &P 500 porfolio as a proxy for he marke porfolio. Hence, in his conex one inerpreaion of he opimal hedge raio is Bea. ince he $ value of each &P 500 fuures conrac is he index value imes $50, he acual number of &P 500 fuures conracs o be wrien is deermined by aking he hedge raio imes he raio of he porfolio $ value divided by he curren value of he index underlying he fuures conrac, he &P 500 in his case. or he example, Number of Conracs 9 = approximaely conracs. $10,968,000 h ˆ = 1.058 46.95 =.16 conracs or ( 934.53 $50) 8 We use monhly reurns because he scale differences in he value of he porfolio and he value of he index. This much easier o scale each of he series and use reurns. 9 ince we calculaed he hedge raio using percenage reurns, he hedge raio does no accoun for he size differenial beween he porfolio and he index. Hence we need o ake his ino accoun when we esimae he number of conracs required. Page 6

Exhibi Je uel and Heaing Oil uures Prices 1997-001 Je uel $/on -,40 lbs. uel Oil 90 day fuures $/meric onne-,05lbs. uel Oil 60 day fuures $/meric onne-,05lbs. Je uel $/on -,40 lbs. uel Oil uures $/meric onne-,05lbs. Price Price Price Price Change Price Change** Jun-97 184.50 81.50 80.93 Jul-97 178.00 8.00 81.43-6.50-0.07 Aug-97 179.00 85.00 84.41 1.00.41 ep-97 174.00 91.50 90.86-5.00 5.86 Oc-97 190.00 96.50 95.8 16.00 4.3 Nov-97 197.50 105.50 104.76 7.50 8.6 Dec-97 186.00 98.00 97.31-11.50-8.19 Jan-98 167.50 79.50 78.94-18.50-19.06 eb-98 151.00 64.00 63.55-16.50-15.95 Mar-98 140.00 68.50 68.0-11.00 4.0 Apr-98 134.00 66.00 65.54-6.00 -.96 May-98 147.50 75.00 74. 13.50 8.47 Jun-98 17.50 61.00 60.57-0.00-14.43 Jul-98 119.00 64.00 63.55-8.50.55 Aug-98 116.00 63.00 6.56-3.00-1.44 ep-98 116.50 57.00 56.60 0.50-6.40 Oc-98 139.00 73.50 7.99.50 15.99 Nov-98 16.00 61.50 61.07-13.00-1.43 Dec-98 101.50 57.00 56.60-4.50-4.90 Jan-99 105.50 59.50 59.08 4.00.08 eb-99 109.50 67.50 67.03 4.00 7.53 Mar-99 108.50 61.00 60.57-1.00-6.93 Apr-99 141.50 64.00 63.55 33.00.55 May-99 157.50 73.00 7.49 16.00 8.49 Jun-99 19.50 65.00 64.55-8.00-8.46 Jul-99 164.00 91.50 90.86 34.50 5.86 Aug-99 17.50 96.00 95.33 8.50 3.83 ep-99 19.00 108.00 107.4 19.50 11.4 Oc-99 08.50 10.50 119.66 16.50 11.66 Nov-99 199.50 13.50 1.64-9.00.14 Dec-99 37.50 18.50 17.60 38.00 4.10 Jan-00 77.00 10.00 119.16 39.50-9.34 eb-00 60.50 18.00 17.10-16.50 7.10 Mar-00 6.00 141.00 140.01 1.50 1.01 Apr-00 64.00 14.00 13.13.00-17.87 May-00 53.00 110.00 109.3-11.00-14.77 Jun-00 58.00 15.00 14.13 5.00 14.13 Jul-00 80.50 140.00 139.0.50 14.0 Aug-00 69.50 116.00 115.19-11.00-4.81 ep-00 330.50 135.50 134.55 61.00 18.55 Oc-00 340.50 153.00 151.93 10.00 16.43 Nov-00 319.50 1.50 147.46-1.00-5.54 Dec-00 33.50 149.50 1.45 13.00-0.05 Jan-01 76.00 100.00 99.30-56.50-50.0 eb-01 4.00 101.00 100.9-34.00 0.9 Mar-01 54.50 119.50 118.66 1.50 17.66 Apr-01 44.50 104.50 103.77-10.00-15.73 May-01 55.00 119.00 118.17 10.50 13.67 Jun-01 58.50 113.00 11.1 3.50-6.79 ** Price change for fuures compares he 60 day price a ime o he 90 day price in ime -1. Page 7

Exhibi 3 Opimal Hedge Raio Je uel and Heaing Oil uures Prices 1997-001 uel Oil Je uel uures $/on -,40 lbs. $/meric onne-,05lbs. Price Change Price Change** (PJ-MJ)(PH-MH) Jun-97 Jul-97-6.50-0.07 0.41 Aug-97 1.00.41 (1.31) ep-97-5.00 5.86 (38.) Oc-97 16.00 4.3 6.85 Nov-97 7.50 8.6 49.36 Dec-97-11.50-8.19 106.47 Jan-98-18.50-19.06 381.47 eb-98-16.50-15.95 87.3 Mar-98-11.00 4.0 (50.71) Apr-98-6.00 -.96.17 May-98 13.50 8.47 101.6 Jun-98-0.00-14.43 310.30 Jul-98-8.50.55 (5.85) Aug-98-3.00-1.44 6.44 ep-98 0.50-6.40 6.64 Oc-98.50 15.99 335.50 Nov-98-13.00-1.43 180.43 Dec-98-4.50-4.90 16.99 Jan-99 4.00.08 5.18 eb-99 4.00 7.53 18.56 Mar-99-1.00-6.93 17.55 Apr-99 33.00.55 80.99 May-99 16.00 8.49 13.06 Jun-99-8.00-8.46 49.11 Jul-99 34.50 5.86 853.03 Aug-99 8.50 3.83 6.79 ep-99 19.50 11.4 0.33 Oc-99 16.50 11.66 174.70 Nov-99-9.00.14 (.75) Dec-99 38.00 4.10 150.3 Jan-00 39.50-9.34 (353.68) eb-00-16.50 7.10 (18.57) Mar-00 1.50 1.01 (0.50) Apr-00.00-17.87 (8.18) May-00-11.00-14.77 184.96 Jun-00 5.00 14.13.93 Jul-00.50 14.0 94.31 Aug-00-11.00-4.81 310.90 ep-00 61.00 18.55 1,104.38 Oc-00 10.00 16.43 139.15 Nov-00-1.00-5.54 14.36 Dec-00 13.00-0.05 (0.8) Jan-01-56.50-50.0,91.39 eb-01-34.00 0.9 (11.1) Mar-01 1.50 17.66 193.81 Apr-01-10.00-15.73 181.31 May-01 10.50 13.67 1.63 Jun-01 3.50-6.79 (13.6) Mean 1.54-0.0 andard Deviaion 0.66 13.40 Covariance 184.1 Correlaion 0.666 Hedge 1.064 Page 8

Exhibi 4 Opimal Hedge Raio Using Regression Analysis UMMARY OUTPUT Regression aisics Muliple R 0.665658193 R quare 0.44310083 Adjused R quare 0.43099436 andard Error 15.74669319 Observaions ANOVA Df M ignificance Regression 1 9075.333 9075.333 36.6003.44E-07 Residual 46 11406.08 47.9583 Toal 47 01.4 Coefficiens andard Error a P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Inercep 1.564717811.7843 0.688441 0.494633-3.0107 6.139708-3.0107 6.139708 X Variable 1 1.064 0.169657 6.04981.44E-07 0.6894 1.367899 0.6894 1.367899 Page 9

Exhibi 5 Equiy Porfolio Hedging Porfolio &P 500 Porfolio &P 500 Reurn index Monhly reurn Monhly reurn (Rp-MP)*(Rs&p-Ms&p) Jan-99 $ 9,78,400 1670.01 eb-99 9,34,100 1730.81-0.00 0.0364-0.0004 Mar-99 8,789,900 168.86-0.01-0.077 0.0013 Apr-99 8,957,00 1763.31 0.0190 0.0478 0.0007 May-99 10,000,100 1847.63 0.1164 0.0478 0.0057 Jun-99 10,044,00 1767.8 0.0044-0.043 0.0000 Jul-99 10,530,300 1888.15 0.04 0.0681 0.0031 Aug-99 10,39,500 1817.7-0.076-0.0375 0.0011 ep-99 9,96,500 184. -0.071 0.0038-0.000 Oc-99 9,574,00 1759.76-0.0390-0.0353 0.0014 Nov-99 9,649,400 1858.86 0.0079 0.0563 0.0001 Dec-99 9,859,500 191.49 0.018 0.0337 0.0006 Jan-00 10,105,300 00.11 0.049 0.040 0.0009 eb-00 9,960,500 1940.4-0.0143-0.0309 0.0005 Mar-00 9,90,000 1901.51-0.0059-0.000 0.000 Apr-00 10,615,600 077.97 0.071 0.098 0.0064 May-00 10,793,400 07.39 0.0167-0.043-0.000 Jun-00 10,700,000 003.45-0.0087-0.0118 0.0001 Jul-00 10,88,400 033.58 0.010 0.0150 0.0001 Aug-00 10,835,900 1991.43 0.0007-0.007 0.0001 ep-00 11,678,800 108.76 0.0778 0.0589 0.0045 Oc-00 11,198,400 199.94-0.0411-0.0549 0.004 Nov-00 11,80,800 1973.7 0.0074-0.0096 0.0000 Dec-00 10,706,400 188.81-0.0509-0.0734 0.0039 Jan-01 11,47,700 1837.36 0.0506 0.0047 0.0004 eb-01 1,363,100 1913.11 0.099 0.041 0.004 Mar-01 11,709,700 1730.91-0.059-0.095 0.0053 Apr-01 10,87,700 1599.35-0.0715-0.0760 0.0056 May-01 1,,900 1769.1 0.166 0.1061 0.0133 Jun-01 1,638,100 1763.87 0.0318-0.0030 0.0000 Jul-01 1,8,900 1731.53-0.0118-0.0183 0.0003 Aug-01 1,343,600 1704.4-0.0116-0.0158 0.000 ep-01 11,79,600 1591.18-0.0446-0.0663 0.0031 Oc-01 9,949,600 1459.33-0.1563-0.089 0.018 Nov-01 10,95,400 154.96 0.1008 0.0450 0.0046 Dec-01 11,739,300 1591. 0.0718 0.0436 0.0031 Jan-0 1,471,300 1618.98 0.064 0.0173 0.001 eb-0 1,99,800 1584.06-0.0138-0.016 0.0003 Mar-0 1,356,900 1600.0 0.0046 0.0101 0.0000 Apr-0 13,085,500 16.3 0.0590 0.0139 0.0009 May-0 1,896,000 1538.65-0.0145-0.0515 0.0010 Jun-0 1,168,500 1476.6-0.0564-0.0405 0.003 Jul-0 11,80,00 1375.87-0.0730-0.0680 0.0051 Aug-0 10,007,000 158. -0.119-0.0855 0.0097 ep-0 10,7,000 1304.85 0.065 0.0371 0.0009 Oc-0 9,407,900 109.59-0.0841-0.0730 0.006 Nov-0 10,080,600 187.13 0.0715 0.0641 0.0045 Dec-0 10,968,000 1337.34 0.0880 0.0390 0.0035 Mean 0.0053-0.0035 and. Deviaion 0.0596 0.0501 Covariance 0.006 Correlaion 0.861 Hedge (Bea) 1.053 Page 10

Exhibi 6 Regression Resuls for Equiy Porfolio Hedging Regression aisics Muliple R 0.863 R quare 0.7435 Adjused R quare 0.7378 andard Error 0.0308 Observaions 47 ANOVA df M ignificance Regression 1 0.140 0.140 130.4440 6.89E-15 Residual 45 0.0 0.0010 Toal 46 0.1668 andard Upper Lower Upper Coefficiens Error a P-value Lower 95% 95% 95.0% 95.0% Inercep 0.0100 0.0045.194 0.0315 0.0009 0.0191 0.0009 0.0191 Bea 1.0581 0.0898 11.41 0.0000 0.8449 1.067 0.8449 1.067 Page 11