Nothing like quitting at the top of your game. I don't blame Henry for getting out. I think he's doing the right thing. I'm sure he'll be in warm climates soon sipping umbrella drinks! What I said is not hindsight. I'm referring to a tool I use to measure volatility and it went ballistic. If I didn't have a quantitative and programming background I would have gotten creamed in 2007/2008. In 2009 I took a beating in the beginning of the year but changed what I was doing because what I was doing wasn't working. I also manage money for clients and as soon as I started losing a number of them jumped ship. Once I started winning again some came back -- classic investor psychology: the sky is falling, sell. At any rate, I trade futures intra-day. I rarely hold overnight and with a strategy such as that you don't have to know much more than the level of volatility -- volatility determines market behavior. It is THEE SINGLE MOST IMPORTANT MARKET MEASUREMENT available to all traders. It is so important it's worth repeating: <b> volatility determines market behavior."</b> If you don't know how to use it you will never be successful. Perhaps the best book to start reading to understand this topic is Natenburg's "Option Volatility and Pricing". The grad school I attended used this book for the Options class I took. It's absolutely priceless. Volatility, Stochastic Processes, Time Series Analysis, etc, etc. These are the tools the pros use. The market is one gigantic random number generator. I can't say that enough. If you don't use tools to analyze randomness you'll never win -- period! Pardon the typos...multi tasking at the moment.... http://www.amazon.com/Option-Volati...UTF8&qid=1353078692&sr=8-1&keywords=natenburg
very interesting there, the1. In the early days of computers, one of the first things people did was try to create a random number generator, because it could be used for testing so many things. It proved to be almost impossibe if not impossible. Many just gave up and and accepted less than perfect results, like 99.99% random. Do you have any tools you use to measure randomness?
OT, agree. A computer cannot generate a perfect random series. I've developed many Monte Carlo simulations and I've yet to find one that can't be mathematically proven as random. Even though I'm a strong proponent of market randomness I'll be the first to tell you the market is not 100% random, which is probably the reason a computer can't generate such. I have tested stock market data and depending on the starting and ending point it may or may not fail the spectral test, which is considered the most powerful test available. So the market can not be asbolutely proven as random or non-random because there are elements of both. I think the real question is: to what degree is the market random? I don't have the answer to that but it's my belief it's highly random because it behaves as a non-stationary or trend-stationary time series. The attached image represents a stationary vs. a non-stationary time series. Which one looks more like the market?
very interesting, so, do you ever bet on past divergence from randomness? in otherwords, do you have a way of measuring when the market is just moving as expected based on your random models and when it is behaving in a non random manner? You're way over my head man, so I'm not sure I am even asking an intelligent question. This all started out years ago when we would just flip a coin, and if it was heads we'd make an uptick on the chart, and if it was tails we would draw a down bar. You don't even need to reply to this silly post. It's just nice to know there is somebody out there that is still thinking about these things.
Boy, what a difficult question. I'm not sure I've even thought of that but what I do throughout different measures of time is plot market data to see what type of curve it generates. Generally, the market will develop a curve that comes close to an F-distribution but depending on the time under consideration you can find n-distributions within those F-distributions and vice versa. Do I view the market as behaving in a non-random manner? I honestly don't know because I'll take a trade on a setup that trades into a double bottom with a lower low but with a + MACD divergence or perhaps a breakout on an ascending triangle. As we discussed earlier we could easily argue that since the market cannot be proven as 100% random I suppose you could argue that I'm trading based on technical analysis rather than statistical analysis even though both models will generate the same type of trade. Or, they may generate opposite trades. Which do you take? That's where the 20 years of experience comes in . I guess the answer to your question is yes. I do deviate from statistical models because the market is not 100% random. That being said, if I were to trade solely on technical analysis I'd be in the 95% category in a hurry. The MS in QF is the best investment I've ever made.
If he's retained enough wealth, maybe she'll stay with him. Otherwise, hypergamy will run its course.
I think the market is chaotic and not random. There is a huge difference between the two. People buy and sell on rules or principles and their psychology. That is definitely not random decisions. Surprisingly most beginners would have better trading results if they did enter randomly. I know of only one methodology based on random entries and it was shown not to work after rigourous testing years ago at trading blox. Just to throw my 2 cents in.
yes, it's funny after a while, the question becomes, "Is there such a thing as random?" You would think the answer would be "Yes" But when they actually tried to generate it, it became darn near impossible. I don't really have all eternity to trade, so I have to make some choices. And for me, in the limited time I have left, I have to go with, "Nothing is random." But if you can find it and prove it, that would sure solve a lot of my problems.
The market thrives on catalysts so therefore it's not random. Due to variable change the market is chaotic but not chaotic enough not to be able to profit. The old saying to make a million dollars trading you lose 9 million and make 10 million. As far as i am concerned the holy grail of trading is a slight edge repeated over and over again. To increase your profits you can move to a lower time frame for higher frequency trading for a faster turn around.
It's surprising how few of you here really "get" the idea of trend following. The poor performance over the last few years has NOTHING to do with JH losing motivation. It's purely a case of market conditions not having been favourable for trend following, i.e. we've had a lot of choppiness and little directional volatility, whereas TF requires the exact opposite. The point is TF does extremely well over a long period of time (10 years) even though big winning years may number only about 3 or 4 in that period. JH has had to close shop because his investors are, surprise surprise, following the herd and acting with recency bias. It's being doing badly recently and so can only ever do badly, right? Wrong, now is the time to be investing in TF lock, stock and barrel. All in. But who is doing it?