Hedging ratio is defined as the ratio of the size of the portfolio taken in hereafters contracts to the size of the exposure ( Hull, 2003, p. 78 ) . Calculating fudging ratio is a necessary measure for the investor who wants to use the hereafter contract to avoid the hazard by fudging in the topographic point market since they should cognize how much contract they should purchase.

In the development history of future hedge, there are three schemes of fudging affecting changeless hedge ratio: the traditional one-to-one hedge, the beta hedge and the minimal discrepancy hedge.

Traditional one-to-one hedge means hedge a unit of a topographic point place one should presume a unit of the opposite place in the hereafters market. Therefore, the optimum hedge ration is equal to one in this scheme. J M Keynes ( 1930 ) and J yokels ( 1939 ) introduced this scheme to investors and they suggest that the intent of take parting in future market of an investor is to use the net income from future trading to do up the possible loss in the topographic point market but non merely to obtain net income. However, this scheme ignores the impact of the footing hazard which means the absence of synchrony of alteration between topographic point market and future market on the hedge. Therefore, it can non assist investor avoid hazard when there is footing hazard between the topographic point and future market.

Compared with the traditional one-to-one scheme which has a hedge ratio equal to 1, beta hedge scheme is a small different. Beta is the correlativity between a stock and index.It suggests that the investors may cognize the beta value of their portfolio and wish to command it utilizing hereafters. For accomplish it, they utilizing Capital plus pricing theoretical account ( CAPM ) to cipher the which is the beta value of the index hereafter with regard to the market portfolio and which is the beta value of the portfolio with regard to the market portfolio. The risk-minimizing hedge ratio for a portfolio is tantamount to: b=/ . However, if the index and the topographic point portfolios change in the same extent which means there is no footing hazard, the consequence got from beta hedge is same with the tradition on-to-one hedge. From Edwards & A ; Ma ( 1992 ) , Bessembinder ( 1992 ) and Antoniou & A ; Holmes ( 1994 ) , beta hedge see the fact that the portfolio hedged in the topographic point market may non fit the portfolio on which the hereafter contract is written. However, Figlewski ( 1984 ) pointed out that the optimum hedge ratio calculated from beta hedge is merely truth when the place is hold until the adulthood of hereafter. Since the hereafters has a more volatility alteration, Peter ( 1986 ) found that the mistake will ensue the volatility of future contract has approximately 20 % more than the existent. Therefore, a over-hedged state of affairs may go on if investor usage beta hedge.

Recognizing the monetary value of topographic point market and hereafters market is parallel but non indistinguishable, Johnson ( 1960 ) and Stein ( 1961 ) argued that the minimal discrepancy hedge which is a portfolio attack is the most sensible scheme. This attack suggests that the substance of hedge is puting a portfolio which includes topographic point market and future market. Hedgers decides the place they take in the topographic point and hereafters market to do the invest risk-minimization or public-service corporation maximising harmonizing to the expect return and discrepancy of expect return. As the same method of beta hedge, minimal discrepancy hedge suppose that the optimum hedge ratio is vary which rely on the correlativity of topographic point and hereafters market and the intent of hedge.

Because the difference of methods of mensurating hazard and pick of public-service corporation map, researches made a great figure of empirical analysis for portfolio hedge which can be categorized as two sorts which are risk-minimizing hedge ratios from the respective of lower limit the hazard and mean-risk hedge ratios from the respective of upper limit the public-service corporation.

Risk-minimizing hedge ratio

Johnson ( 1960 ) foremost advanced the construct of trade good hereafters optimum hedge ratio in the status which is return discrepancy minimisation. In add-on, he represented the equation of mensurating optimum hedge ratio which is called Minimizing Variance Hedge Ratio ( MVHR ) . In his theory, the return of an investor get from the portfolio used to fudge in the hereafter market can be represented as:

( 1 )

where is return of the portfolio keeping in the topographic point and future market between t-1 and T, and intend the return on the topographic point and future market in the same period severally, is hedge ratio. The discrepancy of return of hedged portfolio is represented as

Var ( /Var ( / ) -2Cov ( ) , ( 2 )

where is the information available over the last period. The hedge ratio can be seen as

= ( . ( 3 ) can add basic hazard )

where the value of which minimize the discrepancy of the return is the optimum hedge ratio.

However, Castellino ( 1990 ) said that MVHR can non fulfill the mean-variance model. Therefore it should be under the premise of that investors are decidedly risk averse which can non be existent because some of them is risk appetency for hazard. Ghosh ( 1993 ) argued that this method ignore the historical information and the passable co-integration relation between the topographic point and future monetary value. Therefore, he suggest to use calculate optimum hedge ratio via Vector Auto Regression theoretical account ( VAR ) , Error Correction Model ( EC ) and Fractionally Integrated Error Correction Model ( FIEC ) . Ghosh pointed out that utilizing such theoretical accounts can do full usage of the bing information, better the effectivity of hedge.

However, due to above treatment assumed that the remainders follow normal distribution, have fixed discrepancy and covariance, therefore the optimum hedge ration is calculated as a changeless. But Cecchetti et Al. ( 1988 ) find that the joint distribution of topographic point and hereafters monetary value is changing though clip, the changeless hedge ratio is non suit to minimise hazard. They gave the construct of dynamic hedge and calculated the optimum dynamic hedge ratio with Autoregressive Conditional Heteroskedasticity theoretical account ( ARCH ) . As a consequence they find the optimum hedge ratio showed considerable alterations with time-varying. In add-on, the method of MVHR assumed that the alteration of hereafters follows normal distribution or the public-service corporation map of investor is conelike. But the consequences from tonss of empirical researches show that the fluctuation of hereafters is non under normal distribution and the premise of conic is excessively rough. To get the better of above defect, Cheung, Kwan and Yip ( 1990 ) suggested to utilize Drawn-out Mean-Gini Coefficient ) I“ [ , I» ] ( R [ , t ] ) =-I»COV ( R [ , T ] , ( 1-F ( R [ , t ] ) ) [ I»-1 ] ) as a method to step hazard. Where I» is risk antipathy coefficient, F ( R [ , T ] means the distribution map of return R [ , T ] . The advantage of this method is that the Mean-Gini coefficient has second-order stochastic dominant which means it make non necessitate the premise which MVHR demand. The hedge ratio calculated from this method is called Mean-Extended-Gini hedge ratio ( MEG hedge ratio ) . Kolb & A ; Okunev ( 1992 ) compared the hazard -minizing hedge ratio with extend mean Gini hedge ratio. They found that there was a small difference between these two methods which is for the low degrees of hazard. But extended mean Gini hedge ratio rose with the increasing of hazard antipathy while risk-minimizing hedge ratio kept stable in the same clip. This implies that a equivocator utilizing the exrended mean Gini attack would hold had to continually set their hedge ratio.

De Jong ( 1997 ) examined Generalized Semi-variance method ( GSV ) V [ , I? ] , I» ( R [ , t ] ) = , where I? , I» agencies target return and hazard averse coefficient. F ( R [ , T ] means the distribution map of return R [ , T ] . The definition of hazard in this method is the return which is less than the mark return. The GSV hedge ratio will be measured when minimise the V [ , I? ] , I» ( R [ , t ] ) .

Shalit ( 1995 ) proved that the MEG hedge ratio converge to MVHR hedge ratio if the alteration of future monetary value follows normal distribution. Lien and Tse ( 1998 ) confirmed that GSV hedge ratio is equal to MVHR merely when the topographic point and hereafters monetary value follow articulation normal distribution and the future monetary value follow martingale procedure.

( ??‰e-®e??i??risk minimizing & A ; average viance )

Development of fudging effectivity

Researchers trial fudging effectivity to look into whether the development of weasel-worded point and fudging instrument from theoretical accounts are about countervailing each other. Ederington ( 1979 ) pointed out that the effectivity of risk-minimizing hedge is measured to analyse the impact of hedge on decrease in the discrepancy of net incomes. Since the effectivity calculated by Ederington is merely suited for the risk-minimizing OLS hedge, Gjerde ( 1987 ) , Board & A ; Sutcliffe ( 1991 ) applied another method to calculate effectivity based on returns but non discrepancies. Howard & A ; D ‘ Antonio ( 1987 ) developed a different step of effectivity which is ( c+-r ) / , where degree Celsius is risk free involvement rate, R is expected topographic point return, is standard divergence of topographic point returns and is the swill of the line from the hazard free plus to hedged place. The theory of Howard and D’antonio is simplified by Kuo & A ; Chen ( 1995 ) and extend to use in the multiple plus by Lien ( 1993a ) . However, Lindahl ( 1991 ) argued that the method above is inaccurate when it is utilized to mensurate the portfolio hedges and she invented a two portion effectivity step. Brailsford, Corrigan & A ; Heaney compared three method of measuring hedge effectivity and found the consequence varied.

For the changeless hedge ratio, Figlewski ( 1984a ) pointed out the innovator decision about the fudging effectivity of changeless hedge ratio. He investigated five portfolios which included portions in five US stock indices hedged in the S & A ; P 500 hereafters and take OLS attack to cipher the risk-minimizing hedge ratios which was the portfolio beta. Since fudging effectivity was measured by comparing the standard divergences of returns on the hedged and unhedged portfolios, Figlewski concluded that the fudging effectivity will be limited unless the portfolio of portions which is hedged by investors is similar with the tendency of alteration of the monetary value of the hereafters contracts used for fudging. Hill & A ; Schneeweis ( 1984 ) studied in three US indices which are S & A ; P500, NYSE Composite and VLCI and found the mean hedge effectivity was about 0.8. Junkus & A ; Lee ( 1985 ) besides examines the effectivity with two schemes which are minimal hazard and traditional one-to-one hedge ratio and found that the traditional one-to-one hedge ratio is excessively hapless to increase the hazard for the hedge because the fudging effectivity of each index was really low. For the hazard minimisation scheme, they found that the fudging effectivity measured for S & A ; P500 and NYSE was 0.72 which was much higher than that of VLCI which merely reduced 52 % of the hazard. The different public presentation of indices and besides be found in Figlewski ( 1985 ) , in his consequence hedge in S & A ; P 500, NYSE Composite is more effectual than that of NASDAQ and AMEX Composite.

Model taking

Time-varying

Since there are assorted theoretical accounts to cipher the hedge ratios, there are many research which tried to compared the hedge effectivity of hedge ratios measured from these theoretical accounts. Recent empirical researches focus on the time-varying volatility prevails. Park and Switzer ( 1995 ) suggested that the changeless hedge ratio may non be suited with the changing joint distribution of stock index and hereafters monetary value. In other words, hedge ratios should alter with different conditional distribution between topographic point and hereafters monetary values. Therefore, they suggest to gauging the optimum hedge ratios utilizing time-dependent conditional discrepancy theoretical account such as GARCH.Baillie and Myers ( 1991 ) and Myers ( 1991 ) pointed out that time-varying hedge ratios was superior to constant hedge ratios after analysing the information of US trade good and fiscal hereafters. Kroner and Sultan ( 1993 ) confirmed the consequence of Mayer in currency market. Cheung ( 1993 ) , Teyssiere ( 1997 ) and Dark ( 2005 ) discovered long memory in conditional volatility of stock returns.

The cogent evidence of discrepancy and covariance are clip varying is besides found in the Cu hereafters markets. Hall ( 1991 ) , Figureroal-Ferretti and Gilbert ( 2000 ) used dealing monetary values for aluminum and Cu mental hereafters monetary values to analyze whether the move from manufacturer list pricing to exchange pricing has been associated with an addition in monetary value volatility and they found that the volatility of hereafters monetary value showed an upward tendency over the period in which exchange trading started and exchange monetary value became the established pricing footing. Bunetti and Gilbert ( 2000 ) and Figureroal-Ferretti and Gilbert ( 2004 ) besides found that the hereafters contract monetary values of LME Cooper and Crude oil experienced long memory procedures in volatility.

GARCH

Choosing an appropriate theoretical account is rather important to the truth of the consequence. See the literature above it seems that the theoretical accounts which can gauge a time-varying hedge ratio is better to better the hedge effectivity. Among the theoretical accounts which can supply time-varying hedge ratio, GARCH theoretical account and its extend is most used to analyse hedge in the stock and trade good market. Park & A ; Switzer ( 1995a ) investgatited the S & A ; P 500 for the TSE 35 and found that the hedge effectivity of one-to-one hedge was 73, for OLS was 75 % and bivariate GARCH was 73 % . They concluded that it is better to utilize more sophisticated arrested development processs to mensurate hedge ratio. The same decision can be found in Gagnon & A ; Lypny ( 1997 ) besides, which investigated hebdomadal informations on the TSE 35 with bivariate GARCH theoretical account. Choudhry ( 2003 ) compared the public presentation of difference theoretical accounts in the Nikkei 225, Hang Seng, SPI actioning hebdomadal informations and the consequence stayed stable.

The advantage of GARCH theoretical accounts besides existed in the literature of trade good hereafters. Research showed that the information of trade good hereafters is negatively skewed with heavy kurtosis Chang, Chen and Chen ( 1990 ) , utilizing GARCH theoretical account to mensurating fudging effectivity seems more sensible for these clip series. Bracker and Smith ( 1999 ) and Smith and Bracker ( 2003 ) compared the ability of five theoretical accounts to capture the volatility in Cu hereafter market and realized that the GARCH ranked foremost of the five theoretical accounts, pointed out that GARCH theoretical account provided the best public presentation to suit the Cu hereafters informations in COMEX.In add-on, Watkins and McAleer ( 2002, 2004 ) gave some sentiment after trial non-ferrous metals.

However, Kroner and Sultan ( 1993 ) pointed out that the changeless hedge ratio method was besides simple and accurate after researched in the five currency hereafters from 1985 to 1990. Thomas and Brooks ( 2001 ) besides proved above sentiment, they compared the hedge effectivity between GARCH and OLS hedge ratio and found there was non a important difference. Chakraborty and Barkoulas ( 1999 ) found that the time-varying optimum hedge ratio outperforms the changeless hedge ratio merely in one of five instance they tested. After look intoing the fudging effectivity with the hedge ratio computed by different theoretical accounts, Moosa ( 2003 ) found that there was no important difference for fudging effectivity when he used different theoretical account to cipher the hedge ratio of fudging in Australia and US dollar. Therefore Mossa concluded that compared the choosing of theoretical accounts which were considered as theoretical statements, the success of a hedge depended more on the correlativity between the monetary values of the unhedged place and the hedge instrument.

Hedge continuance

The relationship between hedge continuance and hedge effectivity is another research issue in this paper. There are tonss of literatures which tried attempt to analyse the relationship but the consequences are varied. Junkus & A ; Lee ( 1985 ) pointed out the alteration of fudging effectual for different indices but he found that there is small relationship between contract adulthood and effectivity in the same clip. Malliaris & A ; Urrutia ( 1991 ) verified the influence of hedge continuance to fudge effectivity and the stableness of the estimated hedge ratio utilizing the S & A ; P500 and NYSE Composite indices. After utilizing OLS arrested development theoretical account to gauge the risk-minimmizing hedge ratio which used to mensurate fudging effectivity, they applied Dickey-Fuller and variance-ratio trials to corroborate the hedge ratio they calculated is sensible. The concluding concluded the hedge ratio and effectivity are stable over clip, at least for two hebdomad periods. Deave ( 1994 ) captured the hedge ratio with three method which are traditional one-to-one hedge, the normal OLS attack and a risk-minimizing hedge ratio sing the consequence of adulthood on the risk-free involvement rate. He worked out that the effectivity for one hebdomad hedge of the three sort of method is similar at approximately 98 % and the adulthood of hereafters contract made a small impact on the hedge effectivity.

( ) After analysing the fudging effectivity of five portfolios in different indices of US hedged in the S & A ; P500, Figlewski ( 1984a ) investigated the sensitiveness of his consequence for hedge continuance. He found that the decrease of the hedge continuance from one hebdomad to one twenty-four hours made the hedge effectivity has a important alteration though the consequence is little when he increased of hedge continuance from one hebdomad to four hebdomad. Furthermore, Figlewski ( 1985b ) besides found that the hedges which have hedge continuance less than one hebdomad were non really effectual. He concluded the ground for this state of affairs is because of the usage of ‘stale ‘ monetary values in calculating the five indices. The empirical consequence of fudging effectiveness addition with the fudging continuance can besides found in Lee ( 1994 ) who studied the public presentation of three topographic point portfolios hedged in FTSE 100 hereafters. He test the hedges with one, two and four hebdomads hedge continuance and found that the hedge effectiveness addition with the rise of hedge continuance. For illustration, the hedge effectiveness 44 % for one hebdomad hedge addition by dual to 89 % when the continuance alteration to four hebdomads. Lindahl ( 1992 ) examined the stableness of hedge ratio utilizing the information of MMI and S & A ; P 500. She measured the risk-minimizing hedge ratio utilizing non-overlapping monetary value with one, two and four hebdomad periods for the full investigation period. She found that the hedge ratio and hedge effectivity would increase when the hedge continuance increased. She suggested the continuance consequence may be due to the hedge ratio addition as the continuance hedges bringing were near.

Cross-hedging

If a investor or manufacturer want to do a hedge, the first measure he should take is to choose a hereafter contract to cut down the initial hazard. In by and large the pick is the hereafters contract which can be applied a hedge place with the lowest hazard. However, sometimes there is no hereafter market which can cut down the hazard of topographic point market. Therefore, equivocator should utilize the hereafter market which has the highest correlativity with the topographic point market. There are many literatures about cross-hedging

Hill & A ; Schneeweis ( 1986 ) analyzed the effectivity of international cross hedge of good diversified portfolio. They tested the possible net income for a US investor hedge his portion portfolio in the foreign state and a foreign investor hedge their domestic portfolio utilizing US index hereafters. They found the correlativity for Canada and the UK were over +0.5.