[ Pobierz całość w formacie PDF ]
.On March 19, 2006, the June 2006 contract became the firstexpiration.Because of this expiration, all futures contracts go through a continu-ous change in the way they behave as they progress through their life cycle.The lead contract, or first expiration, is typically the most heavily traded.There are some markets, like the energy markets and the interest rate prod-ucts, which have more volume and liquidity in the more distant expirationsbecause of the way that traders and commercials employ them.Consider an illustration.In January 2006, the September 2006 S&P fu-tures contract was the third deferred delivery and because of this it wasmore thinly traded than the March 2006 first expiration.The majority ofspeculative interest in S&P futures is generally on the first expiration.Be-cause of this lessened focus by traders, the September 2006 contract hadless trading activity.This was evidenced by lower volume and open inter-est.The effect of this lower trading interest on price action was most eas-ily seen as gappy price action and lower volatility, that is, smaller dailyranges.Figures 6.2 and 6.3, respectively, show daily price bars for the March2006 and September 2006 S&P futures contracts each from June 2005to March 2006.The difference is quite apparent.Each chart is markedwith rectangles labeled with their respective expiration position to indi-cate where in the expiration cycle they are from the reference point ofMarch 2006.The S&P September 2006 contract at the start of June 2005 is the fifthexpiration.This is evident from the large number of close-to-open gapsand small range bars.These characteristics progressively diminish as thiscontract moves to the fourth and then third expiration positions.The S&P March 2006 contract at the start of June 2005 is the third ex-piration.This is evident again from the large number of close-to-open gapsand small range bars.These characteristics progressively diminish as thiscontract moves to the second and then completely vanishes as it becomesthe first or spot expiration. c06 JWPR070-Pardo December 18, 2007 14:17 Char Count=122 THE EVALUATION AND OPTIMIZATION OF TRADING STRATEGIESFIGURE 6.2 S&P 09-2006 ContractFIGURE 6.3 S&P 03-2006 Contract c06 JWPR070-Pardo December 18, 2007 14:17 Char Count=The Historical Simulation 123Consider the next stage in the life cycle of the S&P futures December2006 expiration.When the March 2006 contract expired, the September2006 contract moved up from the third expiration to the second expiration.As a result, it attracted a little more trading from speculators and spread-ers.As a result of this increase in volume and speculative activity, its priceaction takes on a slightly different complexion.The final increase in trading volume occurred when the June 2006 con-tract expired and the September 2006 contract became the first expiration,or delivery month.The general pattern of this futures contract expiration life cycle beginswith gappy, low volatility price action because of low volume and openinterest to ends with normal lead contract volatility, volume, and openinterest.It is this finite life and this continuously developing volatility and vol-ume that make the use of actual futures contracts unsuitable for the pur-poses of trading strategy simulations.Typically, speculative trading willtake place in the most liquid contract, which is also most reflective of theprice movement of the underlying cash market.This is usually the first ex-piration.As we will consequently see in the sections that follow, historicalsimulations are best constructed on the basis of a special type of futuresprice history, which features the front expiration on a continuous basis.Also, this process of regular contract expiration and trading interestmoving to the new front expiration creates a need in actual trading to rollopen positions from the expiring contract to the new front contract.For example, assume a long position in the S&P March 2006 contract.Further assume that the strategy chooses to roll positions on the fifth trad-ing day in the spot expiration.This happened to be March 5, 2006.On thatday, the trader simultaneously liquidates his long positions in the March2006 contract and puts on a new and equal position in the S&P June 2006contract.There is typically some difference in the prices between the expiringMarch 2006 contract and the new June 2006 front contract.This is therollover gap.The price gaps that occur at rollover, that is, the transitionfrom the expiring contract to the current active contract, generally do nothave that great of an impact on trading.The strategist must be mindful,however, of the existence of the rollover.There can be circumstances, suchas in the energy futures, which have expirations every month, and wherethe frequency of rollovers might have an impact on real-time performance.Ideally, the simulation should actually include this rollover trade.Inthis way, the impact, if any, of the roll gap, slippage, and commissionswould be accurately represented in the simulation.Most trading simulation software, however, is unable to do accu-rate simulations of rollover trades.The strategist, consequently, must be c06 JWPR070-Pardo December 18, 2007 14:17 Char Count=124 THE EVALUATION AND OPTIMIZATION OF TRADING STRATEGIESmindful and consider whether the roll will have any impact on his strategy.If he determines this to be the case, one simple way to deal with this is toadd in the estimated cost of the roll as additional slippage.In the majority of cases, however, the roll is not much of a factor oneway or another on the performance of most trading strategies.In summary, actual futures price contracts are unsuitable for long-termtesting for two reasons: They are too short and their volume and volatilityare not representative of that which is typically traded.Also, the expira-tion of futures contracts requires the long-term strategy to perform rollovertrades to keep long-term positions intact.A number of data solutions have been offered to solve the problemsthat futures contracts present to the testing of strategies.The majority ofthese solutions involve merging a patchwork of prices from the first expi-rations into some form of continuous contract for the purposes of testing.We will consider these various methods in the next sections.THE CONTINUOUS CONTRACTOne solution to this problem is the continuous contract.This contract is asequential patchwork of successive individual futures contracts.For ex-ample, in January of 2006, the continuous S&P contract had price datafrom the March 2006 contract.In April of 2006, it had price data fromthe June 2006 S&P contract.The continuous contract concatenates pricedata from the most active front contract price expiration into a single pricehistory file.The continuous contract solves two of the three major problems.It canbe as long as required.It has the front expiration contract prices and ac-curately reflects the natural trading vehicle of most speculative traders [ Pobierz całość w formacie PDF ]
zanotowane.pl doc.pisz.pl pdf.pisz.pl odbijak.htw.pl
.On March 19, 2006, the June 2006 contract became the firstexpiration.Because of this expiration, all futures contracts go through a continu-ous change in the way they behave as they progress through their life cycle.The lead contract, or first expiration, is typically the most heavily traded.There are some markets, like the energy markets and the interest rate prod-ucts, which have more volume and liquidity in the more distant expirationsbecause of the way that traders and commercials employ them.Consider an illustration.In January 2006, the September 2006 S&P fu-tures contract was the third deferred delivery and because of this it wasmore thinly traded than the March 2006 first expiration.The majority ofspeculative interest in S&P futures is generally on the first expiration.Be-cause of this lessened focus by traders, the September 2006 contract hadless trading activity.This was evidenced by lower volume and open inter-est.The effect of this lower trading interest on price action was most eas-ily seen as gappy price action and lower volatility, that is, smaller dailyranges.Figures 6.2 and 6.3, respectively, show daily price bars for the March2006 and September 2006 S&P futures contracts each from June 2005to March 2006.The difference is quite apparent.Each chart is markedwith rectangles labeled with their respective expiration position to indi-cate where in the expiration cycle they are from the reference point ofMarch 2006.The S&P September 2006 contract at the start of June 2005 is the fifthexpiration.This is evident from the large number of close-to-open gapsand small range bars.These characteristics progressively diminish as thiscontract moves to the fourth and then third expiration positions.The S&P March 2006 contract at the start of June 2005 is the third ex-piration.This is evident again from the large number of close-to-open gapsand small range bars.These characteristics progressively diminish as thiscontract moves to the second and then completely vanishes as it becomesthe first or spot expiration. c06 JWPR070-Pardo December 18, 2007 14:17 Char Count=122 THE EVALUATION AND OPTIMIZATION OF TRADING STRATEGIESFIGURE 6.2 S&P 09-2006 ContractFIGURE 6.3 S&P 03-2006 Contract c06 JWPR070-Pardo December 18, 2007 14:17 Char Count=The Historical Simulation 123Consider the next stage in the life cycle of the S&P futures December2006 expiration.When the March 2006 contract expired, the September2006 contract moved up from the third expiration to the second expiration.As a result, it attracted a little more trading from speculators and spread-ers.As a result of this increase in volume and speculative activity, its priceaction takes on a slightly different complexion.The final increase in trading volume occurred when the June 2006 con-tract expired and the September 2006 contract became the first expiration,or delivery month.The general pattern of this futures contract expiration life cycle beginswith gappy, low volatility price action because of low volume and openinterest to ends with normal lead contract volatility, volume, and openinterest.It is this finite life and this continuously developing volatility and vol-ume that make the use of actual futures contracts unsuitable for the pur-poses of trading strategy simulations.Typically, speculative trading willtake place in the most liquid contract, which is also most reflective of theprice movement of the underlying cash market.This is usually the first ex-piration.As we will consequently see in the sections that follow, historicalsimulations are best constructed on the basis of a special type of futuresprice history, which features the front expiration on a continuous basis.Also, this process of regular contract expiration and trading interestmoving to the new front expiration creates a need in actual trading to rollopen positions from the expiring contract to the new front contract.For example, assume a long position in the S&P March 2006 contract.Further assume that the strategy chooses to roll positions on the fifth trad-ing day in the spot expiration.This happened to be March 5, 2006.On thatday, the trader simultaneously liquidates his long positions in the March2006 contract and puts on a new and equal position in the S&P June 2006contract.There is typically some difference in the prices between the expiringMarch 2006 contract and the new June 2006 front contract.This is therollover gap.The price gaps that occur at rollover, that is, the transitionfrom the expiring contract to the current active contract, generally do nothave that great of an impact on trading.The strategist must be mindful,however, of the existence of the rollover.There can be circumstances, suchas in the energy futures, which have expirations every month, and wherethe frequency of rollovers might have an impact on real-time performance.Ideally, the simulation should actually include this rollover trade.Inthis way, the impact, if any, of the roll gap, slippage, and commissionswould be accurately represented in the simulation.Most trading simulation software, however, is unable to do accu-rate simulations of rollover trades.The strategist, consequently, must be c06 JWPR070-Pardo December 18, 2007 14:17 Char Count=124 THE EVALUATION AND OPTIMIZATION OF TRADING STRATEGIESmindful and consider whether the roll will have any impact on his strategy.If he determines this to be the case, one simple way to deal with this is toadd in the estimated cost of the roll as additional slippage.In the majority of cases, however, the roll is not much of a factor oneway or another on the performance of most trading strategies.In summary, actual futures price contracts are unsuitable for long-termtesting for two reasons: They are too short and their volume and volatilityare not representative of that which is typically traded.Also, the expira-tion of futures contracts requires the long-term strategy to perform rollovertrades to keep long-term positions intact.A number of data solutions have been offered to solve the problemsthat futures contracts present to the testing of strategies.The majority ofthese solutions involve merging a patchwork of prices from the first expi-rations into some form of continuous contract for the purposes of testing.We will consider these various methods in the next sections.THE CONTINUOUS CONTRACTOne solution to this problem is the continuous contract.This contract is asequential patchwork of successive individual futures contracts.For ex-ample, in January of 2006, the continuous S&P contract had price datafrom the March 2006 contract.In April of 2006, it had price data fromthe June 2006 S&P contract.The continuous contract concatenates pricedata from the most active front contract price expiration into a single pricehistory file.The continuous contract solves two of the three major problems.It canbe as long as required.It has the front expiration contract prices and ac-curately reflects the natural trading vehicle of most speculative traders [ Pobierz całość w formacie PDF ]