CHAPTER 4 第四章
MY ROAD OF TRIALS: 试炼之路
1983–1994
Coming outof my crash, I was so broke I couldn’tmusterenough money to pay for an airplane ticketto Texas to visit aprospectiveclient, even though thefeesI’d earn were many times the cost of the fare—so I didn’t make that trip.Still, I gradually added clients, revenue, and a new team. With time, myupswingsincreased inmagnitudeand mydownswings were both tolerable and educational. I never thought of what I wasdoing as building (or rebuilding) a company; I was just getting the things Ineeded to play my game.
Computerswere among the most valuable things I acquired, because of how they helped methink. Without them, Bridgewater would not have been nearly as successful as itturned out to be.
译文：从惨败中走出来，我几乎破产甚至于我几步没有足够的买一张去田纳西去访问一个潜在客户的机票钱。尽管我可能挣到超过机票的很多倍，所以我没能成行。慢慢的我开始增加客户，收入，以及一个新的团队。随着时间流转，我的势头开始增长巨大而下降趋势开始变得可以接受和赋予教育意义。我从没想过我正在做的事情比如建立（或重建一家公司；我只是在做需要我做的事情。
电脑无疑是我学会的最有价值的技能。因为它能帮助我思考。如果没有电脑，那么桥水基金可能不会取得本应该的成功。
读后感：破产后从新再来，并慢慢积累资金，但这一次和作者刚开始时不同，因为他的起点很高，那些最初的阶段很容易跨过，这就是一个曾经成功的人很容易再次崛起的原因。因为他具备经验，人脉，尤其是失败的经验教会了作者很多很多。
电脑从80年代开始慢慢有了个人电脑，比如苹果，windows电脑，这些最早期的个人电脑可能很贵，但是无疑确实可以帮到作者。
Muster==筹集
----
The first microcomputers (what wouldlater be known as personal computers) had come on the market during the late1970s, and I had been using them as econometricians did, applying statisticsand computing power to economic data to analyze the workings of the economicmachine. As I wrote in a December 1981 article, I believed (and still believe)that “theoretically . . . if there was a computer that couldhold all of the world’s facts and if it was perfectly programmed tomathematically express all of the relationships between all of the world’sparts, the future could be perfectly foretold.”
But I was a long way from doing that.Though my early systems provided valuable insights into where prices wouldeventually reach equilibrium, they hadn’t helped me develop robust tradingstrategies; they just showed me that a particular bet would eventually pay off.For example, I’d run through my analysis and end up with a view that the priceof some commodity should be, say, 75 cents or so. If it was currently 60 cents,I’d know I wanted to buy it,
译文：第一代微型计算机（也就是后来的个人电脑）已经在1970年代末期上市了，我已经再像经济计量学家们那样使用了，应用于统计学和用于运算经济学数据分析。就像我在1981年12月的一篇文章中写的。我相信（现在依旧相信）“理论上讲，如果一台计算机能容纳全世界所有因素，那么他将完美的运算出整个世界关系的完美数学表达式，未来就能完美的预言”。
但是我还有很长的路要走，尽管我的早期系统已经提供了大量有价值的关于价格最终达到平衡见解。他们还不能帮助我发展出简装的交易策略；目前仅仅是提供了一个特定的最终成功模型。举例，经过我的分析，最终得出一些商品价格可能的方向，比如，75美分这样的。如果目前价格是60美分，那么我已经知道要买他了。
读后感：80年代初作者已经运用计算机来辅助计算经济学规律了，比如计算商品价格，并提供决策依据，显然精度还不够，我推理是数据量不够，还有就是算法不够准确，参数设置有问题，这样的精度不足以帮助作者决策。但是方向没错，参数可以再调整，这很可怕，因为到今天已经37年过去了，我相信目前的预测很准确了。但这会不会导致作者丧失了敏锐的个人嗅觉和第六感那？
----
but I wouldn’t be ableto predict that the price would drop to 50 cents before climbing back to 75,and I wouldn’t know when to buy and sell. Rarely, but still too often, thesystem would be dead wrong and I would lose a lot.
“He wholives by the crystal ball is destined to eat ground glass” is a saying I quoteda lot in those days. Between 1979 and 1982, I had eaten enough glass to realizethat what was most important wasn’t knowing the future—it was knowing how toreact appropriately to the information available at each point in time. Inorder to do that, I would have to have a vast store of economic and market datato draw on—and as it happened, I did.
译文：但我依旧不能预测价格是否会跌破50美分，在它攀升到75美分之前，我依旧不知道何时买进何时卖出。非常罕见的，但依旧经常发生，系统可能会彻底错误，而我会损失一大笔钱。
读后感：可以看出模型不够完善哦 ，不能预测下降趋势，系统不能自动给出买进卖出的时间点。还是不能依赖机器啊。
----
From veryearly on, whenever I took a position in the markets, I wrote down the criteriaI used to make my decision. Then, when I closed out a trade, I could reflect onhow well these criteria had worked. It occurred to me that if I wrote thosecriteria into formulas (now more fashionably called algorithms) and then ranhistorical data through them, I could test how well my rules would have workedin the past. Here’s how it worked in practice: I would start out with myintuitions as I always did, but I would express them logically, asdecision-making criteria, and capture them in a systematic way, creating amental map of what I would do in each particular situation. Then I would runhistorical data through the systems to see how my decision would have performedin the past and, depending upon the results, modify the decision rulesappropriately.
译文：从很早期开始，不论我在市场中处于什么位置，我都习惯写下来我做决定的标注。那么，当我结束一笔交易，我就能从哪些标准的工作情况得到反馈。对于我来说如果我将那些标准写入我的公式（现在更流行称之为算法）并通过历史数据运算，我就能通过过去的数据检验我的法则效果如何。实际训练中是这样工作的：就像我一直在做的从直觉开始，但是我将他们逻辑化表达，作为决策制定的标准，并用系统化的方法进行编制，形成在每一个特定条件下的思维导图。接着我就用历史数据来检验系统看看我的决策表现的如何，依据结论，矫正决策规则。
读后感：从初步的直觉到形成逻辑表达式，然后编织成算法，再通过历史数据来验证，反过来矫正决策规则。这就是机器学习的方法。只是作者还没有形成机器的自主学习和训练。不过没关系，现在的机器学习也就是这样了，没有比作者的更高级。当然在数据量和运算能力上不是一个等级了。
----
We testedthe systems going as far back as we could, typically more than a century, inevery country for which we had data, which gave me great perspective on how theeconomic/market machine worked through time and how to bet on it. Doing thishelped educate me and led me to refine my criteria so they were timeless anduniversal. Once Ivettedthose relationships, I could run data through the systems as it flowed at us inreal time and the computer could work just as my brain worked in processing itand making decisions.
译文：我们尽可能的测试系统，典型的都超过一个世纪的数据，每一个国家我们能找到的数据都拿来训练系统，这种训练给了我一个宏大的视角来观察经济/市场及其是如何穿越时间来工作的，如何预测的。这样做同时也教育了我并带我优化我的准则以确保他们的无时间限制和通用性。一旦我诊断那些联系，我就能通过系统运算数据来模拟真实情况，计算机就能像我的大脑一样工作并作出决策。
读后感：雷达里奥老先生不做计算机科学家可惜了，他这一套东东就是机器学习的方法论啊。有理由相信现在桥水基金一定有一套世界先进的机器学习系统。
Vetted==检查 。
----
The resultwas Bridgewater’s original interest rate, stock, currencies, and preciousmetals systems, which we then combined into one system for managing ourportfolio of bets. Our system was like an EKG on the economy’s vital signs; asthey changed, we changed our positions. However, rather than blindly followingthe computer’s recommendations, I would have the computer work in parallel withmy own analysis and then compare the two. When the computer’s decision wasdifferent from mine, I would examine why. Most of the time, it was because Ihad overlooked something. In those cases, the computer taught me. But sometimesI would think about some new criteria my system would’ve missed, so I wouldthen teach the computer. We helped each other. It didn’t take long before thecomputer, with its tremendous processing power, was much more effective thanme. This was great, because it was like having a chess grandmaster helping meplot my moves, except this player operated according to a set of criteria thatI understood and believed were logical, so there was no reason for us to everfundamentally disagree.
译文：结果就是桥水基金的原本的利率、股票、货币，和贵重金属系统-我们都合并到一个系统里用于管理我们的交易业务。我们的系统就像金融行业的晴雨表；系统变化，我们就改变我们的位置，可是，我并没有盲目的听从计算机的指令，我将计算机的结果与我自身的分析并行处理，当计算机和我的结果不一致时，我会检验为什么。大多数时候，都是我错漏了什么事情。那种情况下，我听计算机教的。但有时我会考虑一些新的标准而计算机可能没有的，这时候我来矫正计算机。我们彼此帮助。计算机不会花太长时间，凭借计算机超级的运算能力，比我高效太多了。非常伟大，就像一个想起国际大师再帮我策划每一步，既然这个玩家完全按照我理解和信任的逻辑的一套标准行事，那么我们没有理由有根本性的不合。
读后感：没啥好说的，计算机已经成为作者的左膀右臂了，加上作者的不断矫正和改进，应该无往不胜对吗？？
ECG英文是electrocardiogram的缩写,中文叫规电图或体表电图？
----
The
computer was much better than my brain in “thinking” about many things at once,
and it could do it more precisely, more rapidly, and less emotionally. And,
because it had such a great memory, it could do a better job of compounding my
knowledge and the knowledge of the people I worked with as Bridgewater grew.
Rather than argue about our conclusions, my partners and I would argue about
our different decision-making criteria. Then we resolved our disagreements by
testing the criteria objectively. The rapidly expanding power of computers
during that era was like a constant stream of gifts from the gods to us. I
remember when RadioShack came out with an inexpensive handheld chess computer;
we sent one to each of our clients with the message, “A Systemized Approach
from Bridgewater.” That little computer chess game could kick my ass on level
two out of its nine
levels. It was fun to put it up against each of my clients so they could see
how hard it was to beat computerized decision making.
译文：计算机在多任务思考方面远胜过我的大脑，并能做的更精确，更快捷，没有情绪干扰。并且因为有巨量的存储，在融合、组合各类知识以及我在桥水的雇员越来越多的情况下做的更好。而不是争论结论。我和我的搭档也许会就彼此不同决策标准而争执，然后我们通过系统客观测试来解决问题。那个时代快速膨胀的计算机运算能力就像一条从上帝来的持续的带给我们的礼物流。我还记得RADIOSHACK一家消费电子零售商推出了一款不那么昂贵的手持象棋计算机；我们给每一位客户发送了一则消息“来自桥水一个系统化的应用”那个小计算机象棋游戏9级水平可以轻松击败二级水平的我。通过它与我的客户对战非常有趣，并让他们明白打败计算机化的决策制定系统是多么的难。
读后感：为什么计算机这么厉害,作者还是在1981年遭受了巨大的失败？那时候计算机在干嘛？没有起到作用吗？
Conclusions==结局，结论。
RadioShack==总部设于美国德克萨斯州沃兹堡的 RadioShack Corporation (NYSE: RSH) 是美国信誉最佳的消费电子产品专业零售商。去年已经第二次申请破产了。
----
Of course, we alwayshad the freedom to override the system, which we did less than 2 percent of thetime—mostly totake money
off the tableduring extraordinary events that weren’t programmed, likethe World Trade Center going down on 9/11. While the computer was much betterthan our brains in many ways, it didn’t have the imagination, understanding,and logic that we did. That’s why our brains working with the computer madesuch a great partnership.
These decision-makingsystems were much better than the forecasting systems I’d been using before,mostly because they incorporated our ongoing reactions to developments,allowing us to deal with a wider range of possibilities. They could alsoinclude timing rules. In a January 1987 piece called “Making Money vs. MakingForecasts,” I explained that:
Truthbe known, forecasts aren’t worth very much, and most people who make them don’tmake money in the markets. . . . This is because nothing iscertain and when one overlays the probabilities of all of the various thingsthat affect the future in order to make a forecast, one gets a wide array ofpossibilities with varying probabilities, not one highly probableoutcome. . . . We believe that market movements reflect economicmovements. Economic movements are reflected in economic statistics. By studyingthe relationships between economic statistics and market movements, we’vedeveloped precise rules for identifying important shifts in the economic/marketenvironment and in turn our positions. In other words, rather than forecastingchanges in the economic environment and shifting positions inanticipationof them, wepick up these changes as they’re occurring and move our money around to keep inthose markets which perform best in that environment.
译文：当然我们总是忽视计算机系统的判断-差不多少于2%的时间。尤其是当意外情况没有被编入程序中，比如9/11事件纽约世界贸易中心倒塌。相反计算机在很多方面都胜过人类大脑，它不会拥有人类的推理、逻辑和想象能力，这就是人脑和电脑的天作之合。
那些决策制定系统比我以前用的预测系统强太多了。绝大部分因为计算机融合了持续的改进，允许我们处理大范围的可能性。包括时间性规则。1987年1月piece 叫他“赚钱vs预测”我解释：已知的真相，预测不值钱，大多数人并没有依靠预测在市场上赚到钱。。。因为不确定性，影响未来的事物的可能性被掩盖了，为了做出预测，人们必须在变动的纷繁复杂的可能心中寻找可能，没人能得到靠谱的结果。。。我们相信市场活动反应了经济活动。经济活动反应在经济统计指标中，我们已经发展出一套精确的规则以辨别经济/市场环境下的重要的波动。另一方面，相比在经济环境和变动中预言变化，我们总是拣选那些已经发生的变化，让我们钱跟着在经济环境下表现最好的市场转。
读后感：计算机还不具备人的自主思考和智慧。所以他不具备处理程序之外的能力。而且热钱总是跟着市场热点走，热钱总是向最能发挥效率的地方去。做出预测太难了，影响因素太多，不如根据已知的条件来推测吧。
the World Trade Center
=纽约世界贸易中心
Outcome==结果 后果
输出。
----
Over thelast three decades of building these systems we have incorporated many moretypes of rules that direct every aspect of our trading. Now, as real-time datais released, our computers parse information from over 100 million datasets andgive detailed instructions to other computers in ways that make logical senseto me. If I didn’t have these systems, I’d probably be broke or dead from thestress of trying so hard. We certainly wouldn’t have done as well in the marketsas we have. As you will see later, I am now developing similar systems to helpus make management decisions. I believe one of the most valuable things you cando to improve your decision making is to think through your principles formaking decisions, write them out in both words and computer algorithms,back-test them if possible, and use them on a real-time basis to run inparallel with your brain’s decision making.
But I’m getting ahead
of myself. Let’s go back to 1983.
译文：经过36年建设这些系统我们已经融合了各种指导交易的各种类型的规则。现在，实时数据一释放，我们的计算机就能从超过1亿个数据集中解析信息，给出详细的指令并指导其他计算机做出逻辑判断。如果我没有这些系统，我可能早就破产或者被巨量判断压死了。我们不可能像现在一样做的那么好，就像你后来看到的，我现在开发出了近似的系统帮助我们管理决策。我相信你能改进你的决策制定的最有价值的事情就是运用你的原则决策，用文字和计算机算法写出那些原则，基于实时数据和你的大脑决策并行处理。
但是我已经赶到我自己前头了，让我们回到1983年吧。
读后感：桥水基金的成功不是偶然的，是基于巨量数据和完善的数据处理和分析系统，并在不断的完善改进中。金融业如此，我们也如此，每个人都应该基于数据来判断，而不是单单靠自己的大脑。
本章基本可以看做机器学习和大数据处理的案例，如果你不感兴趣，可以略过，只要知道数据和决策的重要性即可，至于每个人应该怎么做，从收集和个人有关的数据开始吧，然后试着自己推导和分析。我们没有作者那么优秀的大脑和充裕的计算机设备和系统，但是建立初级的属于自己的系统是必须的。
----
读后感：作者留个包袱，估计下一章会揭示真想把 。老实说这一章很啰嗦，是作者的心路历程也是痛苦的探索过程，可以快速略过。
---
我是007李小军，以上是我自己的翻译，也许不会太准确，我能保证也不会偏离作者的意思太远，不至于误导各位，谢谢观看。
007-4478李小军
2018-6-19
觉得好，就点赞，就转发， 谢了！
4617�ԏnؔ