View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by Portsmouth University Research Portal (Pure)
International Research Journal of Finance and Economics
ISSN 1450-2887 Issue 5 (2006)
© EuroJournals Publishing, Inc. 2006
http://www.eurojournals.com/finance.htm

Hedging Effectiveness in Greek Stock Index

Futures Market, 1999-2001

Christos Floros
Department of Economics, University of Portsmouth
Portsmouth Business School, Portsmouth, PO1 3DE, UK
E-Mail: Christos.Floros@port.ac.uk
Tel: +44 (0) 2392 844244

Dimitrios V. Vougas
Department of Economics, University of Wales Swansea
Singleton Park, Swansea, SA2 8PP, UK
Email: D.V.Vougas@swan.ac.uk
Tel: +44 (0) 1792 602102

Abstract

This paper examines hedging effectiveness in Greek stock index futures market. We
focus on various techniques to estimate variance reduction from constant and time-varying
hedge ratios. For both available stock index futures contracts of the Athens Derivatives
Exchange (ADEX), we employ a variety of models to derive and estimate the effectiveness
of hedging. We measure hedging effectiveness using three different methods: (i) the OLS
method, (ii) the method of Ederington (1979), and (iii) the method suggested by Park and
Switzer (1995). In both cases for Greek stock index futures, the hedge ratio from M-
GARCH model provides greater variance reduction, in line with similar findings in the
literature. These findings are helpful to risk managers dealing with Greek stock index
futures.

Keywords: Hedging Effectiveness, Futures, ADEX, OLS, ECM, VECM, M-GARCH.

JEL Classification: G13, G15.

I. Introduction
The basic motivation for hedging is to eliminate/reduce the variability of profits and firm value that
arises from market changes. Hedge effectiveness becomes relevant only when there is a significant
change in the value of the hedged item. A hedge is effective if price movements of the hedged item and
the hedging derivative roughly offset each other. According to Pennings and Meulenberg (1997), a
determinant in explaining the success of financial futures contracts is the hedging effectiveness of
futures contracts.
Ederington (1979) defines hedging effectiveness as the reduction in variance and states that the
objective of a hedge is to minimise risk. Howard and D’Antonio (1984) define hedging effectiveness as
the ratio of excess return per unit of risk of the optimal portfolio of the spot commodity and futures
instrument to the excess return per unit of risk of the portfolio containing the spot position alone (see
also, Pennings and Meulenberg, 1997). Hsin et al. (1994) measure hedging effectiveness by
8 International Research Journal of Finance and Economics - Issue 5 (2006)

considering both risk and returns in hedging. However, all these measures assume that futures contracts
do not introduce risks, an argument which is not correct.
Numerous studies, investigating measures of effectiveness, try to determine to what extent
hedgers are able to reduce cash price risk by using futures contracts. First, Markowitz (1959) measures
hedge effectiveness as the reduction in standard deviation of portfolio returns associated with a hedge.
Then, Ederington1 (1979), following Working (1953, 1962), Johnson (1960) and Stein (1961),
measures hedging effectiveness as the percent reduction in variability. He explains that a hedge is
effective if the R-squared of the OLS regression explaining the data is high, say 90%. But a high R-
squared by itself is not always a reliable indicator of hedging effectiveness. Howard and D’Antonio
(1984) define hedging effectiveness in terms of risk and return. However, the second order conditions
derived by Howard and D’Antonio are incorrect (see, Chang and Shanker, 1987; Satyanarayan, 1998).
In particular, Chang and Shanker (1987) show that the measure of Howard and D’Antonio (1984)
produces inconsistent results. Lindahl (1991) discusses the measures used by Howard and D’Antonio
(1984, 1987) and Chang and Shanker (1987), and states that both measures are not appropriate because
they decrease as basis risk approaches zero. Furthermore, hedging effectiveness has also been
measured by a simple variance or risk-minimisation criterion which indicates whether mean futures are
zero. According to Lypny and Powalla (1998, p. 350), ‘the appropriateness of this criterion depends
on whether mean futures returns are zero; if they are not, hedging may be too expensive’. Finally, most
recent papers use other more advanced econometric methods (i.e. ECM, VECM or BGARCH models)
with or without error correction terms to estimate the hedging performance. In this paper, we examine
whether such methods (ECM, VECM or BGARCH (1,1)) provide better results over the conventional
(OLS) regression in terms of hedging effectiveness. An additional purpose of this article is to
investigate hedging effectiveness in an out-of-sample performance using Greek futures data. This is the
first investigation of hedging effectiveness for Greece.
The paper is organised as follows: Section II provides a detailed literature review, while
Section III shows an overview of methods employed for estimating hedging effectiveness. Section IV
describes the data, and Section V presents empirical results from various methods. Finally, Section VI
concludes the paper and summarises our findings.

II. Detailed Literature Review

Hedging effectiveness has been widely investigated. Most studies focus on the ex post hedging
effectiveness of stock index futures contracts (Figlewski, 1984). Also, little attention has been given to
ex ante2 hedging effectiveness (Malliaris and Urrutia, 1991; Benet, 1990; Holmes, 1995). Next, we
briefly review research papers on hedging effectiveness of financial futures contracts.
Figlewski (1984) studies hedging effectiveness of US stock index futures contracts and
observes that basis risk increases as the duration of the hedging horizon decreases3 . Marmer (1986)
studies hedging effectiveness of Canadian dollar futures over the period July 1981 to September 1984.
Marmer (1986) examines effectiveness of the minimum variance hedge ratio (MVHR) in an ex ante
framework, and shows that usefulness of the MVHR is rather limited. Lasser (1987) considers hedging
effectiveness of Treasury bill and Treasury bond futures contracts. He applies the MVHR on an ex ante
basis, and finds that ‘ex ante hedges generated on the basis of a longer estimation period proved to be
more effective hedges’. Furthermore, Benet (1990) investigates and analyses risk reduction potential on
an ex ante basis with regard to foreign exchange futures contracts. He argues that there is a discrepancy
between hedge ratios on an ex post and ex ante basis, and therefore, his results reveal a more indicative
measure of hedging effectiveness. This is in line with Butterworth and Holmes (2000), who suggest
that ex post hedge ratio leads to a discrepancy between ex post and ex ante effectiveness.

1
Ederington’s (1979) measure has been applied and discussed by Pennings and Meulenberg (1997) and Herbst , Kare and Caples (1989).
2 According to Butterworth and Holmes (2000), hedging effectiveness is truly examined using an ex ante strategy.
3
According to early studies of Ederington (1979), Howard and D’Antonio (1984) and Malliaris and Urrutia (1991), the measures of hedging effectiveness
change with the length of the hedging horizon.
International Research Journal of Finance and Economics - Issue 5 (2006) 9

Holmes (1995) examines ex ante hedging effectiveness of UK index futures contracts (FTSE
100) using data over the period 1984 to 1992. He assumes that the portfolio to be hedged is the one that
underlies the FTSE 100. His results suggest that ‘the introduction of the FTSE-100 futures contract has
given portfolio managers a valuable instrument by which to avoid risk’ (Holmes, p. 59). In addition,
Laws and Thompson (2002) discuss hedging effectiveness of stock index futures on LIFFE, while
Butterworth and Holmes (2000) examine further the ex ante hedging effectiveness of the FTSE 100
and FTSE Mid 250 index futures contracts for a range of portfolios. According to their analysis, ‘the
FTSE 100 contract has been seen to provide the most effective hedge for portfolios dominated by large
capitalisation stocks, and the Mid 250 contract provides the most effective hedge for stocks dominated
by low capitalisation stocks’ (Butterworth and Holmes, 2000; p. 15).
Further, Chang and Shanker (1987) show a new definition of hedging effectiveness following
the model proposed by Howard and D’Antonio (1984, 1987). According to their analysis, Howard and
D’Antonio’s second order conditions do not have to be greater than zero. Also, Jong et al. (1997) apply
three models for hedging effectiveness of futures to measure the effectiveness of currency futures: (i)
the minimum-variance model of Ederington (1979), (ii) the a-t model of Fishburn (1977), and (iii) the
Sharpe ratio model of Howard and D’Antonio (1984, 1987). Their results indicate that hedges are only
effective for the minimum-variance and the a-t models. Brailsford, Corrigan and Heaney (2000)
discuss several techniques of measures of hedging effectiveness using the Australian All Ordinaries
share price index futures contract. Their results show that there is an impact from selection of the
measure of hedge effectiveness to the assessment of hedged portfolios.
In addition, Chou, Denis and Lee (1996) compare hedging performance using Japan’s NSA and
NSA index futures with different time intervals. They find that the conventional hedge outperforms the
error-correction hedge over the in-sample period. However, the out-of-sample period evaluates better
hedging strategies. Chou, Denis and Lee (1996) also find that the error-correction model outperforms
the conventional model over the out-of-sample period.
Park and Switzer (1995) examine hedging effectiveness for three types of stock index futures:
(i) S&P 500, (ii) MMI futures and (iii) Toronto 35 index futures. Their results show that the bivariate
GARCH estimation improves hedging performance over the conventional constant (OLS) hedging
strategy. Furthermore, Bera, Garcia and Roh (1997) use a bivariate GARCH model and a random
coefficient autoregressive (RCAR) model to examine hedging performance of spot and futures prices
of corn and soyabeans. They find that the diagonal vech presentation of BGARCH model provides the
largest variance reduction of the return portfolio.
Lypny and Powalla (1998) examine hedging effectiveness of German stock index DAX futures
using a bivariate GARCH (1,1) model and an error-correction of mean returns. Empirical results
confirm that the dynamic model is superior to models with constant hedge with or without error-
correction means. This is in accordance with Kroner and Sultan (1993). Kroner and Sultan (1993)
argue that a bivariate error correction model with GARCH error structure leads to more effective
hedges than the conventional (OLS) method. More recently, Kavussanos and Nomikos (2000) show
that a GARCH-X model outperforms all other hedges, while constant hedge ratio provides greater
variance reduction over the sample. However, out-of-sample results report that an ECM-GARCH-X
model outperforms alternative hedging strategies. Also, Yang (2001) shows that M-GARCH dynamic
hedge ratios provide the greatest degree of variance reduction.

III. Methodology
Different measures of hedging effectiveness include: the early measure of Markowitz (1959), the
method of Ederington (1979), the measures by Howard and D’Antonio (1984, 1987), and Lindahl’s
(1991) measure.
An early measure of hedge effectiveness is introduced by Markowitz (1959). He measures
hedging effectiveness in accordance to the reduction of the standard deviation of portfolio returns
10 International Research Journal of Finance and Economics - Issue 5 (2006)

associated with a hedge. In this case, the greater the reduction in risk, the greater the hedging
effectiveness.
Ederington (1979) states that hedging effectiveness is equal to R-squared of the OLS
regression:
∆S t = c + b∆Ft + u t (1)
where S t and Ft are logged spot and futures prices at time period t, respectively, and ut is the error
term from OLS estimation. ∆S t and ∆Ft represent spot and futures price changes.
Ederington (1979) shows that a hedge is effective if the R-squared of the regression line
explaining the data is high. In other words, the higher the R-squared, the greater the effectiveness of
the minimum-variance hedge.
However, effectiveness of the minimum-variance hedge can be determined by examining the
percentage of risk reduced by the hedge (Ederington, 1979; Yang, 2001). Hence, the measure of
hedging effectiveness is also defined as the ratio of the variance of the unhedged position minus the
variance of the hedged position, over the variance of the unhedged position4 :
Var (u ) − Var (h)
E=
Var (u ) (2)
where
Var (u ) = σ S2
Var (h) = σ S2 + h 2σ F2 − 2hσ S , F
Ru = S t +1 − S t
Rh = ( S t +1 − S t ) − h( Ft +1 − Ft ) (3)
Var (u ) and Var (h) represent variance of unhedged and hedged positions, respectively, while
σ S , σ F are standard deviations of the spot and futures prices, respectively. The hedge ratio is defined
as the value of h and σ S , F represents the covariability of the spot and futures price.
Further, Park and Switzer (1995) and Kavussanos and Nomikos (2000) measure the variance of
the hedged returns to the portfolios by evaluating Var (∆S t − ht ∆Ft ) , where ht is the computed hedge
ratio. In this case, the variance reduction (i.e. HE) is calculated as:
σ 2 (Unhedged ) − σ 2 ( Hedged )
HE = (4)
σ 2 (Unhedged )

IV. Data
The data employed in this paper comprise 525 daily observations on the FTSE/ASE-20 stock index and
stock index futures contract (August 1999-August 2001) and 415 daily observations on the FTSE/ASE
Mid 40 stock index and stock index futures contract (January 2000-August 2001). Closing prices for
spot indices were obtained from Datastream, and closing futures prices were obtained from the official
web page of the Athens Derivatives Exchange (www.adex.ase.gr).
Focusing on the above periods, (i) we test the hypotheses using data from the early stage of the
ADEX (started its official operation on 27 August 1999), and (ii) we investigate whether the
hypotheses exist after the dramatic rise of Athens Stock Exchange (ASE) stock prices5 .
The FTSE/ASE-20 comprises 20 Greek companies, quoted on the Athens Stock Exchange
(ASE), with the largest market capitalisation (blue chips), while the FTSE/ASE Mid 40 comprises 40

4 Similarly, Butterworth and Holmes (2000) examine the ex ante hedging effectiveness of the FTSE 100 and FTSE Mid 250 index futures contracts using
the minimum variance strategy of Johnson (1960).
5
The Athens Stock Exchange (ASE), an important European emerging equity market, experienced a dramatic rise of stock prices between the years 1998-
1999, followed then by an equally dramatic fall.
International Research Journal of Finance and Economics - Issue 5 (2006) 11

mid-capitalisation Greek companies. Futures contracts are quoted on the Athens Derivatives Exchange
(ADEX). The price of a futures contract is measured in index points multiplied by the contract
multiplier, which is 5 Euros for the FTSE/ASE-20 contract and 10 Euros for the FTSE/ASE Mid 40
contract. There are four delivery months: March, June, September and December. Trading takes place
in the 3 nearest delivery months, although volume in the far contract is very small. Both futures
contracts are cash-settled and marked to market on the last trading day, which is the third Friday in the
delivery (expiration) month at 14:30 Athens time.

V. Empirical Results
In this section, we compare measures of hedging effectiveness using several types of hedging models.
Since selection of the hedge effectiveness measure has a considerable impact on the assessment of
hedged portfolios, we measure hedging effectiveness of Greek stock index futures markets using
Model (1), Model (2) and Model (4)6 . To do so, the hedge ratios obtained from the methods of OLS,
ECM, VECM and BGARCH (1,1) 7 are considered (see Floros and Vougas, 2004). The main question
in this section is whether hedge ratios calculated from several methods generate better results in terms
of hedging effectiveness.
First, Ederington (1979) argues that hedging effectiveness is equal to R-squared of the OLS
regression (1). A hedge is effective if the R-squared of the OLS regression (1) is very high. Table 1
reports R-squared results for FTSE/ASE-20 and FTSE/ASE Mid 40. The R-sq. value for FTSE/ASE-
20 is much higher than that of the FTSE/ASE Mid 40. Also, the fact that the R-sq. of FTSE/ASE-20
index is greater than 0.80 indicates that the hedge is effective. This is also in line with the main
conclusion from previous studies that futures contracts perform well when R-sq. is between 0.80 to
0.99.

Table 1: In-Sample Hedging Effectiveness (R-Squared)

FTSE/ASE-20 FTSE/ASE Mid 40

R- sq. = 0.847163 (= 84.7163%) R- sq. = 0.717509 (= 71.7509%)
Notes: Model: ∆S t = c + b∆Ft + u t
Next, we measure hedging effectiveness of the FTSE/ASE-20 and FTSE/ASE Mid 40 using
expressions (2) and (3). That is, we estimate the variance of the unhedged (i.e. Var(u)) and the hedged
portfolios (i.e. Var(h)), and then calculate the hedging effectiveness, E. To do so, we take into
consideration hedge ratios8 estimated from the OLS regression, ECM, VECM and mean value of the
selected BGARCH model. Table 2 reports (in-sample) results of hedging effectiveness from Model 2.
It is clear that the FTSE/ASE-20 contract produces the most effective hedges. Also, for both contracts
the OLS hedge ratio provides greater variance reduction. This is in contrast with recent papers (Yang,
2001), which show that the hedge ratio, calculated from the conventional regression model, does not
perform well in terms of variance reduction (hedging effectiveness).

6
We select to measure hedging effectiveness by using Models (1), (2) and (4) because both the Howard and D’Antonio (1984, 1987) and Lindahl (1991)
measures require the risk-free rate of return series.
7
We consider a restricted version of the Bivariate BEKK model by Engle and Kroner (1995), i.e. a Bivariate cointegration model with GARCH (1,1)
error structure.
8
The hedge ratios are estimated as follows: OLS 0.9160, ECM 0.9123, VECM 0.9129 and BGARCH (1,1) 0.9235 (FTSE/ASE-20), and OLS 0.7033,
ECM 0.7150, VECM 0.7204 and BGARCH (1,1) 0.7542 (FTSE/ASE Mid 40).
12 International Research Journal of Finance and Economics - Issue 5 (2006)
Table 2: In-Sample Hedging Effectiveness (Model 2)

METHOD FTSE/ASE-20 FTSE/ASE Mid 40

OLS hedge 0.850121 = 85.0121% 0.71575 = 71.5751%
ECM hedge 84.9866% 71.554%
VECM hedge 84.8711% 71.531%
BGARCH hedge 84.9875% 71.196%
Notes:
Var (u ) − Var (h)
Model: E=
Var (u )
Further, following Park and Switzer (1995) and Kavussanos and Nomikos (2000), we evaluate
the variance of returns over the sample using Model (4). In other words, hedging effectiveness is now
investigated by calculating Var (∆S t − ht ∆Ft ) , where ht is the computed hedge ratio from the OLS,
ECM, VECM and BGARCH models. In addition, we calculate percentage variance reductions of the
unhedged position (i.e. Var(S t )) over the other four hedges. Therefore, the variance of hedged
positions is compared to the variance of unhedged position. The variance reduction (i.e. hedging
effectiveness, HE) is now calculated as:
σ 2 (Unhedged ) − σ 2 (OLS , ECM ,VECM , orBGARCH )
HE = (5)
σ 2 (Unhedged )
According to Park and Switzer (1995) and Kavussanos and Nomikos (2000), the larger the
reduction in the unhedged variance, the higher the degree of hedging effectiveness. The variances are
reported in Table 3, while the in-sample variance reductions are presented in Table 4.

Table 3: In-sample variances of Returns

METHOD FTSE/ASE-20 FTSE/ASE Mid 40

OLS hedge 0.000059536 0.000157653
ECM hedge 0.000059552 0.0001577536
VECM hedge 0.000059552 0.0001578792
BGARCH hedge 0.0000581711 0.0001474767
Unhedged 0.000389 0.000556

Table 4: Percentage variance reductions (Model 5): In-sample

METHOD FTSE/ASE-20 FTSE/ASE Mid 40

OLS hedge 84.695% 71.645%
ECM hedge 84.691% 71.627%
VECM hedge 84.691% 71.604%
BGARCH hedge 85.045% 73.475%
σ 2 (Unhedged ) − σ 2 (OLS , ECM ,VECM , orBGARCH )
Notes:Model: HE =
σ 2 (Unhedged )

Our results show that the BGARCH (1,1) hedge ratio provides greater variance reduction than
the other models. Thus, the hedge ratio obtained from the Bivariate cointegration GARCH(1,1) model
generates better results in terms of hedging effectiveness. In particular, for FTSE/ASE-20, the hedge
ratio provides greater benefits from hedging as it significantly reduces the risk of price movements (i.e.
the BGARCH (1,1) method reduces the variance by 85%). This is consistent with recent research
papers (Park and Switzer, 1995; Kavussanos and Nomikos, 2000; Yang, 2001).
However, the more reliable measure of hedging effectiveness is the hedging performance of
the post-sample periods. Since investors need to predict all about the future, we use an out-of-sample
(post-sample) performance measure, which represents a way to evaluate effectiveness of hedge ratios.
For this, we collect 100 observations of the sample (i.e. from April 2001 to end of August 2001). For
International Research Journal of Finance and Economics - Issue 5 (2006) 13

OLS, ECM and VECM hedging models, estimated hedge ratios are used for out-of-sample period. For
BGARCH (1,1) model, we perform one-step ahead forecasts of covariance and the variance. Hence,
the forecasted hedge ratio is the one-period forecast of the conditional covariance divided by the one-
period forecast of the conditional variance. These out-of-sample estimated variances of returns are
reported in Table 5, while the out-of-sample percentage variance reductions are reported in Table 6.
Our results show that the hedge ratio9 obtained from OLS and ECM generates better results in terms of
hedging effectiveness. Both methods reduce variance by almost 90% and 85% for FTSE/ASE-20 and
FTSE/ASE Mid 40, respectively. Also, we calculate Root Mean Squared Errors, Mean Absolute Errors
and Mean Absolute Percent Errors for post-sample forecasts from the OLS method and Error
Correction Model. Our results are presented in Figure 1 (FTSE/ASE-20) and Figure 2 (FTSE/ASE Mid
40). It can be observed that, Root Mean Squared Errors show that the error correction model (ECM)
outperforms the OLS (conventional) model in both indices. Therefore, in this case the error correction
model (ECM) is superior to the conventional model. This is consistent with Chou et al. (1996).

Table 5: Out-of-sample variances of Returns

METHOD FTSE/ASE-20 FTSE/ASE Mid 40

OLS hedge 0.000034187409 0.000049773025
ECM hedge 0.000034187409 0.000049773025
VECM hedge 0.000034292736 0.000049787136
BGARCH hedge 0.000034621456 0.000049928356
Unhedged 0.000375274384 0.000327429025

Table 6: Percentage variance reductions (Model 5): Out-of-sample

METHOD FTSE/ASE-20 FTSE/ASE Mid 40

OLS hedge 90.890% 84.798%
ECM hedge 90.890% 84.798%
VECM hedge 90.861% 84.794%
BGARCH hedge 90.774% 84.751%
Notes:
σ 2 (Unhedged ) − σ 2 (OLS , ECM ,VECM , orBGARCH )
Model: HE =
σ 2 (Unhedged )

9
The hedge ratio for the out-of-sample period is estimated as following: OLS 0.829832, ECM 0.831213, VECM 0.844990 and BGARCH (1,1) 0.872479
(FTSE/ASE-20), and OLS 0.787543, ECM 0.784684, VECM 0.794749 and BGARCH (1,1) 0.769032 (FTSE/ASE Mid 40).
14 International Research Journal of Finance and Economics - Issue 5 (2006)
Figure 1: FORECASTING (FTSE/ASE-20)

.08
Forecast: OLS METHOD (FTSE/ASE-20)
.06 Forecast sample: 426 525
Included observations: 100
.04
Root Mean Squared Error 0.005817
.02 Mean Absolute Error 0.004570
Mean Abs. Percent Error 156.6240
Theil Inequality Coefficient 0.153856
.00
Bias Proportion 0.000000
Variance Proportion 0.023872
-.02
Covariance Proportion 0.976128
-.04

-.06
450 475 500 525

FORECAST: OLS METHOD

.08
Forecast: ECM METHOD (FTSE/ASE-20)
.06 Forecast sample: 426 525
Included observations: 100
.04
Root Mean Squared Error 0.005044
.02 Mean Absolute Error 0.003981
Mean Abs. Percent Error 116.0238
Theil Inequality Coefficient 0.132665
.00
Bias Proportion 0.000017
Variance Proportion 0.018627
-.02
Covariance Proportion 0.981355
-.04

-.06
450 475 500 525

FORECAST: ECM METHOD

International Research Journal of Finance and Economics - Issue 5 (2006) 15
Figure 2: FORECASTING (FTSE/ASE Mid 40)

.06
Forecast: OLS METHOD (FTSE/ASE MID 40)
.04 Forecast sample: 316 415
Included observations: 100
.02
Root Mean Squared Error 0.007019
.00 Mean Absolute Error 0.005134
Mean Abs. Percent Error 190.5715
Theil Inequality Coefficient 0.201896
-.02
Bias Proportion 0.000000
Variance Proportion 0.041195
-.04
Covariance Proportion 0.958805
-.06

-.08
325 350 375 400

FORECAST: OLS METHOD

.08
Forecast: ECM METHOD (FTSE/ASE MID 40)
.06 Forecast sample: 316 415
Included observations: 100
.04
Root Mean Squared Error 0.006273
.02
Mean Absolute Error 0.004741
.00 Mean Abs. Percent Error 117.1410
Theil Inequality Coefficient 0.178873
-.02 Bias Proportion 0.000003
Variance Proportion 0.031915
-.04 Covariance Proportion 0.968082

-.06

-.08
325 350 375 400

FORECAST: ECM METHOD

VI. Conclusions
On theoretical grounds, a hedge is effective if price movements of the hedged item and the hedging
derivative roughly offset each other. Ederington (1979) defines hedging effectiveness as the reduction
in variance and states that the objective of a hedge is to minimise risk, while Howard and D’Antonio
(1984) define hedging effectiveness as the ratio of excess return per unit of risk (of the optimal
portfolio of the spot commodity) and the futures instrument to the excess return per unit of risk (of the
portfolio containing the spot position alone). Furthermore, Hsin et al. (1994) measure hedging
effectiveness by considering both risk and returns in hedging.
In this paper, we investigate hedging effectiveness of Greek stock index futures contracts
(FTSE/ASE-20 and FTSE/ASE Mid 40). We compare several techniques of measuring hedging
effectiveness. In particular, we measure hedging effectiveness of Greek stock index futures markets
using three different methods: (i) the OLS model, (ii) the measure of hedging effectiveness which is
defined as the ratio of the variance of the unhedged position minus the variance of the hedged position,
over the variance of the unhedged position, and (iii) the method suggested, and applied, by Park and
Switzer (1995) and Kavussanos and Nomikos (2000).
16 International Research Journal of Finance and Economics - Issue 5 (2006)

The primary objective of this paper is to examine whether specific hedge ratios (calculated
from different methods) generate better results in terms of hedging effectiveness. Our hedge ratios
obtained from the methods of OLS, ECM, VECM and BGARCH (1,1). First, results from OLS suggest
that the R-sq. value of FTSE/ASE-20 is much higher than that of FTSE/ASE Mid 40. Therefore, the
FTSE/ASE-20 index provides larger risk reduction. However, the ECM method indicates that the
FTSE/ASE-20 contract produces most effective hedges. Following Park and Switzer (1995) and
Kavussanos and Nomikos (2000), we show that the BGARCH (1,1) hedge ratio provides even greater
variance reduction than the ones from other models. So, the hedge ratio obtained from the Bivariate
cointegration GARCH(1,1) model generates better results in terms of hedging effectiveness. This is in
accordance with Park and Switzer (1995), Kavussanos and Nomikos (2000) and Yang (2001).
Finally, we measure hedging effectiveness by considering hedging performance for post-
sample periods. Using forecasting statistics for OLS and ECM, we find that the root mean squared
error of ECM is lower than that of OLS for both indices. Hence, the ECM outperforms the OLS
(conventional) model, and therefore, the error correction model (ECM) is superior to the conventional
model. This is consistent with Chou et al. (1996).
International Research Journal of Finance and Economics - Issue 5 (2006) 17

References
[1] Benet, B. A. (1990), Ex-ante reduction of foreign exchange exposure via hedge ratio
adjustment, Review of Futures Markets, 9, 418-435.
[2] Bera, A. K., Garcia, P., and Roh, J-S. (1997), Estimation of time-varying hedge ratios for corn
and soybeans: BGARCH and Random Coefficient Approaches, Paper, The Office for Futures
and Options Research, Department of Agricultural and Consumer Economics, University of
Illinois at Urbana-Champaign.
[3] Brailsford, T. J., Corrigan, K., and Heaney, R. A. (2000), A comparison of measures of hedging
effectiveness: A case study using the Australian All ordinaries share price index futures
contract, Working Paper, Department of Commerce, Australian National University.
[4] Butterworth, D., and Holmes, P. (2000), Ex ante hedging effectiveness of UK stock index
futures contracts: evidence for the FTSE 100 and FTSE Mid 250 contracts, European Financial
Management, 6, 441-457.
[5] Chang, J., and Shanker, L. (1987), A risk-return measure of hedging effectiveness: a comment,
Journal of Financial and Quantitative Analysis, 22, 373-376.
[6] Chou, W. L., Denis, K. K. F. and Lee, C. F. (1996), Hedging with the Nikkei index futures: the
convential model versus the error correction model, Quarterly Review of Economics and
Finance, 36, 495-505.
[7] Ederington, L. H. (1979), The hedging performance of the new futures markets, Journal of
Finance, 34, 157-170.
[8] Engle, R. F., and Kroner, K. F. (1995), Multivariate simultaneous generalized ARCH,
Econometric Theory, 11, 122-150.
[9] Figlewski, S. (1984), Hedging performance and basis risk in stock index futures, Journal of
Finance, 39, 657-669.
[10] Fishburn, P. C. (1977), Mean-risk analysis with risk associated with below-target returns,
American Economic Review, 67, 116-126.
[11] Floros, C., and Vougas, D. V. (2004), Hedge ratios in Greek stock index futures market,
Applied Financial Economics, 14, 1125-1136.
[12] Herbst, A. F., Kare, D. D, and Caples, S. C. (1989), Hedging effectiveness and minimum risk
hedge ratios in the presence of autocorrelation: foreign currency futures, Journal of Futures
Markets, 9, 185-197.
[13] Holmes, P. (1995), Ex ante hedge ratios and the hedging effectiveness of the FTSE-100 stock
index futures contract, Applied Economics Letters, 2, 56-59.
[14] Howard, C., and D’Antonio, L. (1987), A risk-return measure of hedging effectiveness: a reply,
Journal of Financial and Quantitative Analysis, 22, 377-381.
[15] Howard, C., and D’Antonio, L. (1984), A risk-return measure of hedging effectiveness, Journal
of Financial and Quantitative Analysis, 19, 101-111.
[16] Hsin, C.-W., Kuo, J., and Lee, C. W. (1994), A new measure to compare the hedging
effectiveness of foreign currency futures versus options, Journal of Futures Markets, 14, 685-
707.
[17] Johnson, L. (1960), The theory of hedging and speculation in commodity futures, Review of
Economic Studies, 27, 139-151.
[18] Jong, A., Roon, F., and Veld, C. (1997), Out-of-sample hedging effectiveness of currency
futures for alternative models and hedging strategies, Journal of Futures markets, 17, 817-837.
[19] Kavussanos, M. G., and Nomikos, N. K. (2000), Constant vs. time-varying hedge ratios and
hedging efficiency in the BIFFEX market, Transportation Research Part E, 36, 229-248.
[20] Kroner, K. F., and Sultan, J. (1993), Time varying distributions and dynamic hedging with
foreign currency futures, Journal of Financial and Quantitative Analysis, 28, 535-551.
[21] Lasser, D. J. (1987), A measure of ex ante hedging effectiveness for the treasury bill treasury
bond futures markets, Review of Futures Markets, 6, 278-295.
18 International Research Journal of Finance and Economics - Issue 5 (2006)

[22] Laws, J., and Thompson, J. (2002), Hedging effectiveness of stock index futures, Paper, Centre
for International Banking, Economics and Finance (CIBEF), Liverpool John Moores
University.
[23] Lindahl, M. (1991), Risk-return hedging effectiveness measures for stock index futures,
Journal of Futures markets, 11, 399-409.
[24] Lypny, G., and Powalla, M. (1998), The hedging effectiveness of DAX futures, European
Journal of Finance, 4, 345-355.
[25] Malliaris, A. G., and Urrutia, J. (1991), Tests of random walk of hedge ratios and measures of
hedging effectiveness for stock indexes and foreign currencies, Journal of Futures Markets, 11,
55-68.
[26] Markowitz, H. M. (1959), Portfolio selection: efficient diversification of investments, John
Wiley and Sons, Inc., New York.
[27] Marmer, H. (1986), Portfolio model of hedging with Canadian dollar futures: a framework for
analysis, Journal of Futures Markets, 6, 83-92.
[28] Park, T. H., and Switzer, L. N. (1995), Time-varying distributions and the optimal hedge ratios
for stock index futures, Applied Financial Economics, 5, 131-137.
[29] Pennings, J. M. E., and Meulenberg, M. T. G. (1997), Hedging efficiency: a futures exchange
management approach, Journal of Futures Markets, 17, 599-615.
[30] Satyanarayan, S. (1998), A note on a risk-return measure of hedging effectiveness, Journal of
Futures Markets, 18, 867-870.
[31] Stein, J. L. (1961), The simultaneous determination of spot and futures prices, American
Economic Review, 51, 1012-1025.
[32] Working, H. (1962), New concepts concerning futures markets and prices, American Economic
Review, 52, 431-459.
[33] Working, H. (1953), Futures trading and hedging, American Economic Review, 43, 314-343.
[34] Yang, W. (2001), M-GARCH Hedge Ratios and Hedging Effectiveness in Australian Futures
Markets, Paper, School of Finance and Business Economics, Edith Cowan University.