【原文推送】克罗谈投资策略（16）

Everyone Has a System

If you had scrutinized the apparel of the gentleman sitting alongside my desk a number of years ago - the one trading his head off in beans - you would have noticed something distinctly curious. The guy always wore a brown necktie - which may not have been so unusual, but it was always the same brown necktie. Since he was otherwise elegantly dressed, you might well have asked, Why the same brown necktie every day?

如果是多年前，你能够看见我座位旁边坐着一位衣冠楚楚的绅士——他疯狂地交易黄豆。你一定会注意到这个奇怪的地方。这位仁兄总是系一条棕色领带。由于他的穿着一向优雅，你难免会问：为什么天天系同一条领带？

His answer was simple: The brown necktie was his Soybean Trading System. His Soybean Trading System? Exactly so - and for a period of several months, this otherwise intelligent and perfectly rational gentleman, this corporate officer of a major drug company, was totally convinced that his Soybean Trading System actually worked.

他的回答很简单：这条棕色领带是他的“黄豆交易系统”。他的“黄豆交易系统”？没搞错吧？没错——几个月的时间内，这位本来很聪明，十分理性的仁兄，某个制药公司的主管，完全证实他的“黄豆交易系统”绝对管用。

The guy - let’s call him Marvin - had never actually traded commodities before, but at his club he overheard two “experienced traders” exchanging bullish gossip about soybean futures. That was all he needed to hear. The very next day, he walked up to the Merrill Lynch office where I was a freshman account executive, asked for a broker who knew how to read the commodity board (that turned out to be me), and opened his account. After I showed him how to read the grain prices on the big commodity board (these were the days before desk-top screens) plus some requisite logistics such as hours of trading, size of contracts, commissions, margins, and margin calls, he plunked down a check for $5,000 and plunged into 10M bushels (two contracts) of May soybean futures. But then, the most implausible and remarkable thing happened. He somehow lucked into the exact beginning of the renowned 1961 soybean bull market. We all watched incredulously as this total neophyte actually lurched from 10M to 225M beans (from 2 to 45 contracts) and from $10,000 to over $80,000 in equity.

这位仁兄——我们暂且叫他马文吧——以前从没交易过商品，但是有一天他在俱乐部里听到两个“经验丰富的交易者”交换黄豆期货利好心得，他觉得这样就够了。第二天，他来到美林公司，当时我还是那里新进的业务员。他说要找一个看得懂商品看板的人帮他的忙（这件事就落在了我的身上），他开了户。我带他到商品大看板前，教他怎么看谷物价格（那时还没有电脑屏幕可以看），并让他了解一些必要的开户问题，例如交易时间、合约的大小、手续费、保证金和保证金追缴程序。听完，他就掏出5000元的支票，马上买入1000万蒲式耳（2份合约）的5月黄豆期货。接着，令人难以置信的事情发生了。他很幸运，正好赶上了十分有名的1961年黄豆多头市场。我们用不敢相信的眼神，看着这位刚出道的新手从1000万蒲式耳做到2亿2500万蒲式耳（2份到45份合约），投入资金增加到10000元，再增加到80000元。

And so it happened. The bean market had been limit down for three consecutive sessions (it was just a 10 cent limit back then). Marvin was losing $22,500 per day - probably half a year’s pay - and he was feeling the hot breath of the margin clerk breathing down his back. Then he appeared in our office sporting his good luck necktie. It had become his good luck necktie when he once came from behind to win a big bridge tournament while wearing it. As luck would have it (although you could never get him to concede it was due to luck), after having been locked limit-down since the opening, the market staged a prodigious rally in the final 15 minutes of the session. It soared from limit-down to unchanged to, would you believe, LIMIT-UP. That evening Marvin treated himself, good luck necktie and all, to his first limo ride, from our midtown office to his apartment in the Bronx. Needless to say, from that day forward, regardless of whatever else he wore, his good luck brown necktie invariably encircled his freely perspiring neck. A curious trading system to be sure, but he was absolutely convinced that it worked, and that’s what really counted for him. Regrettably, it stopped working shortly after his equity surpassed the $80,000 mark, but that’s a different story altogether.

但是天有不测风云，后来黄豆市场竟然连续3天跌停（当时跌停幅度只有10分）。马文每天亏掉22500元——可能是半年的薪水——开始感觉到催缴保证金的压力。然后他打着那条幸运的领带，出现在我们的办公室。有一次他打着这条领带，在某次重要的桥牌锦标赛中后来居上，赢得冠军，从此这条领带就成了他的幸运符。就是那么幸运（但你绝对没办法要他承认事情纯属巧合），那一天开盘跌停后，临收盘前最后15分钟价格惊人反弹，跌停打开，涨到平盘，再涨到——信不信由你——涨停。那天晚上，马文好好款待自己和那条领带，坐他新买的高级轿车，从我们在城里的办公室开回布郎克斯的公寓。不用说，从此以后，不管他穿什么衣服，那条幸运棕色领带总是一成不变地绑在他那挥汗如雨的脖子上。这当然是个很奇怪的交易系统，但是当事人绝对相信它管用，对他来说，这才是最重要的。不过很不幸的，在他投入的资金超过80000元后不久，幸运领带失灵了，不过那是后话了。

What isn’t a different story is that we now live in an advanced, technological world, so that our trading systems can depend more on logic, mathematics, and computer programs than on good luck brown neckties.

不一样的是，我们现在生活在先进的科技世界中，我们的交易系统可以依赖更多合乎逻辑的程序，运用数学方法，利用电脑程序来运算，而不必靠幸运的棕色领带。

At the forefront of this technology is the current crop of powerful microcomputers. In lightning speed, they can do studies and calculations that would have been unthinkable just a few years ago.

这些科技中打头阵的是目前功能强大的电脑，它们能在电光石火之间，研究和计算出几年前还难以想象的东西。

Everywhere I go and in just about every phone conversation and piece of correspondence, the focus seems to revolve around computer systems, software, on-line quotes, tick-by-tick charts, and so forth. The computer allows even the neophyte trader, after minimum hours studying the workings of the equipment and the operation of the software, to begin testing existing systems and to develop some of his own. He can combine interesting segments of different systems and try to come out with something that will fulfill his needs; that is, assist him in profiting from his trading activities - winning more on profitable positions, losing less on adverse ones.

我每到一个地方，每一通电话，每一封信，谈论的焦点似乎总是围绕着电脑交易系统、软件、线上报价和分时图等东西。新手只要花几个小时，研究电脑的交易方法和软件的运作方式，就能开始测试现有的系统，发展自己的系统。他可以把不同系统的有趣部分结合起来，设计出满足自己需求的东西：也就是帮他从交易中获利——赚钱的仓位赚的更多，亏钱的仓位亏的更少。

So now, everyone has some sort of trading system. But was this always so? Here’s some commentary about speculation and trading systems that I came across recently:

所以说，现在每个人都有某种形式的交易系统。但是它们都能算做交易系统？以下是我最近偶然看到的评论，主要谈投机和交易系统的问题：

The reader ought to thoroughly understand, however, that there is no royal road to speculation. Given all the conditions of the problem, and profits could be ciphered out with the accuracy of a mathematical demonstration. But the unknown quantities are the stumbling blocks of system mongers...Successful operators...cannot explain, even to themselves, yet they know when a stock is a good purchase, and when its price is running too high. None of them would allow that a system is possible, or could perfect their combinations, except by an unconscious use of this very impossibility.

但是，读者应该彻底了解，投机没有所谓的金光大道。并不是知道了问题的所有状况，利润就可以用数学公式精确地表达。会有很多无法预测的事，成为系统使用者的绊脚石……成功的交易者自己都说不清一支股票什么时候适合买入，什么时候适合卖出。他们没有一个人知道系统有可能，或可以使他们的综合选择更为准确，而只是无意识地运用这些很不可能的事情。

How easy it would be to fill a book with these figures! The charts teem with them. There is not a stock... which does not indicate...alternations of hope and fear in wide percentages and with unbounded opportunities for speculators to buy with confidence of profit. Brokers tell us that about one in a hundred buy in this manner. The 99 mean to do so. They make their calculations, add up, subtract, wander hither and thither for points, try this system and that theory, are wise to the extent of their wisdom, and come forth from their ventures shorn of all their golden fleece.

把这些数字填在本子上是多么容易的一件事！图表到处都是。每只股票都是希望和恐惧交织，投资者有无限的机会买入某支股票，深怀信心一定赚钱。经纪人告诉我们，大约100个人当中有1个人是用这种方式买入。剩下的99个人不是这样子，他们会做计算，加减，四处找重点，试试这个系统又试试那个理论，冒险之后，身上的钱都亏完了。

Do these words sound familiar to you? Have you read it in some Wall Street book or in one of the market advisories? You may have read something similar to this but, as Alphonse Karr once said, “The more things change, the more they remain the same.” In fact, it’s not likely that you’ve read this particular Wall Street quote recently - it was written by James K. Medbury in 1870 (James K. Medbury, Men and Mysteries of Wall Street (Burlington, Fraser Publishing, 1870, reprint 1968).).

这些话你听起来是不是很熟悉？你是不是会在华尔街上某本书里面看到过？或者是和某个市场专家说的一样？你可能看过类似的文句，不过，就像阿方斯·卡尔所说的：“事情变化越多，它们越保持原貌。”事实上，最近你不可能看到这段华尔街特别的话，因为那是詹姆士·K·梅德贝利在1870年所写的。（詹姆士·K·梅德贝利，《人类和华尔街的神秘》，1870年，1968年重印。）

If traders were experimenting with systems back in the last century, they are more than ever doing it in the current one. In fact, it is probably true that just about every serious trader has, at one time or another, experimented with some type of trading system to improve the timing of trade entries, keep himself in a profitable position longer, or get out of a losing position sooner. All three of these objectives are important, and overall dollar success largely depends on the trader being able to pursue and attain these objectives.

如果交易者曾经在上个世纪试验过系统，现在就会认真用了。事实上，这件事可能是真的：几乎每一个认真的交易者都会试验过某种交易系统，改善进出时机，希望待在赚钱的仓位中尽可能长的时间，或者尽早退出亏钱的仓位。这三个目标都很重要，整体结算下来能不能赚钱，主要是看交易者有没有能力追求和达成这些目标。

The universal appeal of a good long-term, trend-following system is that positions are generally established in the direction of the trend reasonably near the onset of the move. A good system tends to keep you in the position so long as the market continues to move favorably and to stop you out, possibly reversing as well, when the trend reverses. The rub here lies in the phrase, “tends to keep you in the position” because, as all systems traders have discovered, it is frustratingly difficult to fine tune the market stops to stay aboard so long as the market is moving favorably, but (and this is the key “but”) to stop and possibly reverse at just the right moment. In the real world of tension-filled, white-knuckled trading, stop placement will be an imperfect art at best. The stop will either be too close - and you will keep getting whipsawed on minor technical corrections - or too far - and you end up taking whopping big losses or else giving back most of your paper profits when the trend reverses. The solution to the problem of fine tuning the stops is probably the toughest part of successful systems design and is clearly the area that receives the maximum focus in new research and studies.

一个良好跟踪趋势的长期系统最引人的地方，是仓位通常建立在趋势就要发动的起始点，而且是往趋势行进的方向建立。一个良好的系统，会让你在市场继续对你有利的情况下，保持原有的仓位不动，一旦不利，又会用止损的方式让你退出，如果趋势反转，它也会叫你反向交易。这里的困难之处在于“尽量保持仓位不动”，因为所有的系统交易者都发现，很难调整好市场止损点，以保证在市场仍对你有利的情况下能继续待在车上，但是（重要的就是这个“但是”两个字）能在恰到好处的时刻，透过止损点把你拉出市场，可能的话还要反向交易。这是很难的。真实的世界充满了紧张气氛，设定止损点只能说是一项艺术，没办法像科学那样精确。止损点可能设得太接近——只要小小的技术性回调，你就会被洗出市场——或者距离太远——最后使你抱着巨大的赤字黯然出场，或是在趋势反转的时候，吐回大部分的帐面利润。成功交易系统在设计的时候，最困难的地方可能就是要想办法解决止损点要怎么微调到恰到好处的地方，这也是目前研究交易系统的人最关心的问题。

Another major problem with trend-following systems is that, during periods of broad sideways price movement - and these are far more prevalent than established, dynamic trends - the system follower is invariably buying on rallies and selling on reactions. These whipsaw losses are an inevitable part of trend-following trading. The operator needs have the patience and financial stamina to sit through a succession of whipsaw losses while waiting for the big move that will provide the big profits. It takes considerable patience and discipline to stick with a system that is broadly whipsawing you in and out of positions. However, our experience has shown that, once you have embarked on a viable trend-following approach, you will do far better sticking with it and following it to the letter than trying to second-guess and “improve” it.

趋势跟踪系统的另一个大问题，是价格在宽广的区间内横向盘整时——这种情况比激烈的既定趋势更常见——根据系统交易的人总是会在反弹时买入，回调时卖出。这种洗盘的损失，是顺势交易不可避免的一部分，交易者必须有耐心，财力也要雄厚，渡过一连串的洗盘损失，等候大行情的到来，大赚一笔。系统一直让你进出不断，并且发生损失时，你确实需要有很大的耐心且守纪律，才能依照系统所说的去做。但是我们的经验显示，一旦你坚决采用某套可行的趋势跟踪方法，遵行它的成果，会比你一再怀疑它的能力，不断想办法“改进”它要好得多。

As trading systems go, the moving average approach is clearly the oldest and most basic of all methods. In its simplest form, a moving average (m.a.) is the sum of X consecutive closes divided by X. For example, you obtain a nine-day simple m.a. by adding the closing prices for the last nine days and dividing the sum by nine. Perhaps the most popular simple m.a. combinations are the 5- versus 20-day, and the 4- versus 9- versus 18-day.

谈到交易系统，我们就想到移动平均法。这个方法显然是所有方法中最古老和最基本的一种。最简单的移动平均数是拿X个连续收盘价之和除以X。比如说，如果你要得九天的简单移动平均数，那你可以把过去九天的收盘价加起来再除以九。最简单的移动平均数组合，可能是5天对20天，以及4天对9天对18天。

The “versus” enters the picture because traders have discovered through years of trial, error, and testing that a “crossover” technique captures the maximum advantage of a moving average. There are, essentially, two ways to play these simple systems, and they can amaze traders by sometimes outperforming more complicated and elaborate ones. You can take trades when the closing price crosses your m.a. line; for example, using a 19-day approach, you would buy when the price closes above the 19-day m.a. and sell (possibly reversing as well) when the price closes below the 19-day m.a. This simple system, however, offers less flexibility than the second approach, which uses a dual cross over; for example, a 5-20-day crossover, where you buy when the short-term line (the 5-day) closes above the long-term line (20-day) and liquidate the long, and go short as well, when the 5-day line closes below the 20-day (In the 4-9-18-day strategy, you close longs when the 4-day closes below the 9-day, go short when the 9-day closes below the 18-day, and so forth.).

这里用了“对”字，是因为交易者经过多年来的摸索，犯下错误和试验之后，发现“交叉”或“穿越”技巧，最能发挥移动平均线的效果。本质上，有两种方法运用这些简单的系统，而且有些时候移动平均线的表现竟比其它更为复杂和精细的系统要好。当收盘价穿越你的移动平均线时，你就可以进行交易；比如说，使用19天移动平均线时，价格往上穿越19天移动平均线时，你可以买入，当价格穿越19天移动平均线往下时，你可以卖出（甚至反向交易）。但是这个简单的系统，比第二种方法缺少弹性。第二种方法是利用双交叉法：比如说，以5天和20天移动平均线交叉法来运用时，当短期移动平均线（5天）收盘价往上穿越长期移动平均线（20天）时就买入，反之，如果5天移动平均线向下穿越20天移动平均线，则卖出，甚至做空。（如果使用4天对9天和18天的均线策略，当4天移动平均线向下穿过9天移动平均线时，平掉多头的仓位；当9天移动平均线向下穿过18天移动平均线时，开始做空。以此类推。）

These basic strategies will surely get you aboard early in the trend, but you will just as surely encounter lots of premature moves coupled with whipsaw stop losses. However, the payoff arrives when the system gets you and keeps you in a position for an extended move.

这些基本的策略当然能使你及早投身于趋势之中，但是你也一样肯定会遭遇很多言之过早的波动，且因上下波动止损出场而损失。但是，如果趋势持续相当长的时间，你就会尝到甜头。

Serious systems traders tend to get considerably more involved with moving averages. Some use so-called weighted moving averages, which give greater importance to recent than to older price action. For example, a 15-day weighted average might factor the most recent close at 15, the previous day at 14, and so forth, till you get to the oldest of the 15 days, which would be factored at 1. Then you would multiply the most recent close by its factor (15), the previous close by 14, and so forth, dividing the sum by the total of the factors, 120 in this case, in order to get the pertinent average. Other operators utilize exponentially smoothed averages, which incorporate a potentially infinite time span via more complex calculations. Such approaches clearly require the use of a sophisticated calculator or, as a more practical arrangement, a computer with customized software.

认真的系统交易者往往会更深入运用移动平均线。有些人利用所谓的加权移动平均数，这种方法对最近价格的重视甚于以往的价格。比如说，15天的加权移动平均线可能会给最近5天的收盘价权数15，前一天的权数给14，以下依此类推，直到15天前的收盘价给1为止。然后再把最近5天的收盘价乘以15，前一天的收盘价乘以14，以下依此类推，然后再把总数除以权数的和（这个例子中是120），目的是为了得到中肯的平均值。还有些交易者使用指数平滑平均数。这个方法是利用更为复杂的计算，跨越一个可能无限长的时段。这种方法显然需要用到计算器，更务实的方法则是装一台电脑，里面有专为这个目的而设计的软件。

For any moving average system - regardless of its complexity - a critical question concerns the number of days in your average and whether it should be optimized (tailored) to each distinct commodity. In this regard, some of the best technical research has been done by technical analysts Frank Hochheimer and Dave Barker. Hochheimer tested a broad array of averages, from 3 to 70 days, on each of 13 different futures for the 1970-76 period. His results clearly indicated that there was no single “best” universal combination. His best simple moving average (closing price going through a moving average value), which projected best profits, looks like this:

对任何移动平均系统来说——不管是简单的，还是复杂的——一个十分要紧的问题是：移动平均数要用多少天？以及每一种不同的商品，是不是要用最适当的天数（专为那种商品而选定的天数）？谈到这点，我们不妨看看技术分析师弗兰克·霍克海马和戴夫·巴克所做十分杰出的研究。霍克海马测试过1970到1976年13种不同的期货，每一种都用十分广泛的平均数，从3天到70天不等来测试。他的结果很明显地指出，没有单一“最佳”的通用组合。他的最佳简单移动平均数（收盘价穿越某个移动平均值）估计出的最佳利润如下：

Please note that these are hypothetical trades, done on the basis of an after-the-fact calculation. Real-time results are unlikely to show such profits. Also note the low ratio of profits to total trades, from .201 to .366 - typical of systems and formula trading. Hochheimer went on to test the best combinations of linearly weighted moving averages, exponentially smoothed moving averages, and finally simple versus exponentially smoothed versus linearly weighted moving averages. The results of this research have been published in the 1978 Commodity Year Book (Commodity Research Bureau, New York).

请注意这些是假设性的交易，是根据事后的计算而得。即时的交易成果不可能有上述的利润。同时也请注意利润对总交易次数的比率偏低，从.201到.366不等——这是系统和公式交易的典型结果。霍克海马又测试了线性加权移动平均线，指数平滑移动平均线，最后则相互之间比较，比较了简单移动平均线，指数平滑平均移动线和线性加权移动平均线。研究结果发表在1978年的商品年鉴中（商品研究局，纽约）。

For those traders interested in going beyond the simple, closing price versus a single moving average crossover, the dual moving average crossover approach would be the next step. With this technique, you calculate both a short- and a long-term moving average, say an 8- and a 35-day average. You would buy when the 8-day crossed above the 35-day and sell when the reverse occurred. Here again, Hochheimer did some excellent research in the testing of optimum crossover periods, using 20 different commodities for the years 1970-79. Some of his optimum combinations are:

收盘价对一条移动平均线的方法很简单，有人可能觉得不满意，下一个研究领域是双重移动平均线交叉法。利用这种方法时，必须计算短期和长期的移动平均值，比如说，8天和35天平均值。8天线穿越35天线往上时买入，相反的情况发生时则卖出。同样的，霍克海马做了十分杰出的研究，利用1970——1979年的20种不同商品，测试最适当的交叉组合，一些最适当的组合情况如下：

Another analyst who did some fine work in systems testing was Dave Barker, who tested the 5-and-20-day dual m.a. crossover system (without optimization) versus an optimized dual m.a. crossover system for the 1975-80 period. Not surprisingly, the optimized version consistently outperformed the straight 5-and-20-day version, and a partial list of Barker’s best combinations follows:

戴夫·巴克是另一个系统测试方面做了认真研究工作的分析师。他针对1975年——1980年的市场，对比了5天和20天双重移动平均线交叉系统（没有优化）和优化后的双重移动平均线交叉系统。结果不奇怪，优化后的系统比没有优化的系统表现要好。巴克的部分最佳组合情况如下：

There are lots of other ways to use moving averages, some of them focusing on the momentum of the move. A London-based currency analyst of my acquaintance uses moving averages to both identify trends and to time trades. His overall focus is quite long-term, utilizing one-month (21-day) and three-month (63-day) averages in his technical work. As a start, he compares daily the price of the 21-day m.a. versus the price of the respective future 63 days ago, getting a sense of the momentum of the recent move. Then he examines the slope of the 63-day m.a., discerning a buy signal when the curve changes slope from down to up and a sell signal when the opposite occurs. Because this is quite a long-term approach, there are relatively few trades. This analyst told me that he made just five D-mark trades in 3.5 years and claimed that his overall results were profitable. I’ve spoken to other analysts as well, many of whom went beyond moving average crossover analysis, focused instead on the slope of the m.a. lines, and took their signals on the change in slope of their various lines. However, I have not seen any real-time results from these methods.

移动平均线的运用还有很多种方法，有些专注在移动的动量。我认识伦敦一位外汇分析师便是利用移动平均线来找趋势和进出时机。他的整个焦点放得相当长，技术研究工作是利用1个月（21天）和3个月（63天）的移动平均线。开始的时候，他每天比较21天移动平均价格与相对应的63天线，用来寻找最近趋势的动力。接着，他会检查63天线的斜度，如果曲线斜度由下跌转为上涨，就是买入信号，反之则是卖出信号。由于这是一个相当长期的方法，所以交易次数相当少。这位分析师告诉我，三年半的时间内只做了5次德国马克的交易，而且整个结算下来有获利。我也跟其他的分析师谈过，其中很多人都超越了移动平均线交叉的分析，重心改在移动平均线的斜度上，从各条线的斜度变化找出信号。不过，我没看过这些方法产生立即的成果。

（本章未完）

相关阅读：

【原文推送】克罗谈投资策略（01）

【原文推送】克罗谈投资策略（02）

【原文推送】克罗谈投资策略（03）

【原文推送】克罗谈投资策略（04）

【原文推送】克罗谈投资策略（05）

【原文推送】克罗谈投资策略（06）

【原文推送】克罗谈投资策略（07）

【原文推送】克罗谈投资策略（08）

【原文推送】克罗谈投资策略（09）

【原文推送】克罗谈投资策略（10）

【原文推送】克罗谈投资策略（11）

【原文推送】克罗谈投资策略（12）

【原文推送】克罗谈投资策略（13）

【原文推送】克罗谈投资策略（14）

【原文推送】克罗谈投资策略（15）

长按图片点击“识别图中二维码”关注本混混的公众平台~

................................................................................