by MrBill » Fri 11 Nov 2005, 04:00:37
$this->bbcode_second_pass_quote('CARVER', 'T')hanks for the lesson MrBill. I'm also curious if there are some great stories/myths of people that claim to have found the perfect (Technical) Analysis, one that defies the laws of the market. The point where greed takes over common sense, and even some of the properly educated professionals start to believe, because they want to believe it so much. Such stories exist in the field of physics and computer science/math for example. There must be one here too? (I vaguely remember hearing something about some belgium trader that claimed to have figured out the market?)
Sure. Now that commodities are
back in the headlines you hear outrageous claims all the time. Here in Cyprus there was one
broker advertising on the radio with claims like,
Investor Morse, he's not a financial advisor, he's a money manager. 80% of his predictions this year have been correct. His overall capital growth has been over 60%. etc. What were his predictions? The sun will rise tomorrow? I think I will have lunch at noon? I am sure I will make more money than my clients? 60% capital appreciation? What from $100 to $160? Where's the baseline?

I am pretty sure he has not legit because I don't hear his radio commericals anymore, so someone must have complained to the CySEC about him?
In any case, people do regression analysis and back testing all the time and find models that fit past price action and therefore they think their model works. A lot of academic egg heads for instance. However, mostly their models optimize past performance and are poor indicators of future price movements. It depends whether you are in a bull market or a bear market. Whether the market is trending or sideways and choppy. Whether there is high volatility or rather quiet conditions. It is hard to build a model to fit all market conditions.
Usually these
academics fail to take into considerations transaction costs and bid/offer spreads. The most famous example being the Noble prize winners who
invented the Black-Scholes option pricing model (actually the mathmetics was borrowed from
a dust/heat particle expansion model from engineering) but ran Long Term Credit Corporation into bankruptcy when they tried to manage a hedge fund based on their models. As they put it,
they were going to vacuum up nickels from the bottom of the swimming pool that no one knew were even there.
Take a simple set of moving averages. If you would have used a weekly chart and a 30-day moving average you could have easily caught the entire move from $15 to $70 with a buy & hold strategy. You would be quite rich. In hindsight it is quite plain. However, when you look at week to week volatility you would have had moves of up to 10% during those weeks and it would have taken brass balls to hang on. For instance the top of the channel was at $70.85, but the bottom just broke under $60, so that is a move of 15% before the uptrend was broken. But in hindsight a brilliant trade.
If you use short term moving averages. Say an hourly chart and a 14-period average, you will catch every single move. However, you may have to buy/sell, buy/sell, buy/sell 2-3 times per day. The transaction costs will be very high. Also you will miss the peaks and the lows, so you may end up just paying away spread during the day. But, if you follow this strategy religiously, you will catch any larger, trending moves.
So, yes, there are strategies that work, but they all have their trade-offs.

The organized state is a wonderful invention whereby everyone can live at someone else's expense.