Market Timing: Should You Attempt It?

Top 285. Nov 10, 2003 2:57 PM

» Normxxx - Re: All About Market Timing and Day trading on the edge

In response to message posted by radiodude:

radiodude-- The only thing wrong with day trading is that 90 - 95% of the people doing it, shouldn't have. Bear markets are really much better for day trading if you know what you are doing, since the volatility is usually much higher. In recent markets, which have grudgingly traded through a 10 point range, if that, there is hardly any action to take advantage of.

Since most day traders during the Bull did not short, that's like trading with one hand tied behind your back.

I don't know if Day Trading on the Edge was a ripoff or not (it probably was, since if Masonson was good at it, he would have done it, not written about it), but Market Timing is easily verified by independent sources of the stats. Unfortunately, you are not likely to get rich market timing, but it can help keep you from being killed if you can't pick stocks like Kirk.

-- posted by Normxxx

Top 286. Nov 10, 2003 3:09 PM

» radiodude - Re: Re: Book: All About Market Timing

In response to message posted by Normxxx:

All the 200 day MA does is Low Pass Filter the daily noise on the recent market data to show a smoother curve based on what has occurred over the last 200 days. Since it is causal, it has no ability to predict the future, even when comparing it to the current value of the market.

I always wonder why chartists use 200 days and not some other value. I would guess that 200 works better for the data-miners.

Backtested, the 200 day MA would not have gotten one out before the recent market fall, but rahter well after it started. One must keep in mind that it is causal, and equal weight is given to each data point for 200 days, so not only do you get out well after the fall has started, but you get in well after the recovery has started.

When the market goes down again (whenever that is) you will see that the 200 MA gets you out well after the peak of the market, and in after a runup is well underway.

-- posted by radiodude

Top 287. Nov 10, 2003 3:31 PM

» Normxxx - Re: Re: Re: Book: All About Market Timing

In response to message posted by radiodude:

I think if you run the numbers, you will find that crude though it is, it did better than buy and hold over the last 5 years or so. 200 is used because of historical reasons and avoids the need to recalculate a lot of past stats. I think someone on this board was singing the praises of the 210dma. At least 2 of my "circuit breakers" use 150 dma.

However, this will be the first year out of the last 6 that the Seasonal Timing System has not beaten the pants off of buy and hold. In fact, STS produced only 1 down year (but it stll beat the indexes), whereas the indexes turned in 3 losing years.

Check it out. I think you will find that in most Bull years, nothing much happens over the summer; in Bear years, most of the grief happens in the summer (to include September and possibly May and/or October). You might want to check out Sy Harding's book, "Riding the Bear."

-- posted by Normxxx

Top 288. Nov 10, 2003 3:40 PM

» radiodude - Re: Re: Re: Re: Book: All About Market Timing

In response to message posted by Normxxx:

Norm, I assume you know how the 200dma is calculated. And knowing this you see that it is just a low pass filter --right?

Logically, how could this tell you anything about the future direction of the market since the 200dma is causal and subject to a delay since the equally weighted data goes back 200 days?

Again, you will get out after things start to go down, and in after things have started to go up. --- just the nature of Low Pass Filters.

-- posted by radiodude

Top 289. Nov 10, 2003 3:52 PM

» Normxxx - Re: Re: Re: Re: Re: Book: All About Market Timing

In response to message posted by radiodude:

There are no stats which can predict the future. Even if I assume the sun has risen every day for the past 4.5 billion years or so, I have no basis for making any prediction about whether it will rise tomorrow or not! By definition, tomorrow's data point cannot be a member of any set on which I can calculate probabilities.

However, my philosophy is, if it works, don't knock it. If I can (post hoc) find a reasonable explanation, that's nice, but unnecessary. All of science works on similar lines. We use a historical set of data to come to some kind of conclusion, then if we can reasonably predict a future outcome, we assume we have a useful relationship.

Outside of physics, there are darned few things that fit a causal model.

All of the trend following models assume that the market has a tendency to trend. In fact, while short term fluctuations can be shown to be random, the longer term fluctuations do show a tendency to trend. The market is not exactly a "random walk."

-- posted by Normxxx

Top 290. Nov 10, 2003 4:01 PM

» radiodude - Re: Re: Re: Re: Re: Re: Book: All About Market Timing

In response to message posted by Normxxx:

Norm, You may not understand the math behind the 200dma but you need to know that if it is used as a tool for getting in and out of the market, then you will get out after things have started to go down, and you will get back in after things have started to go up --- in many cases the 200dma can harm returns since you will be getting in after a runup and out after things have gone down a fair amount. It's all in the math.

-- posted by radiodude

Top 291. Nov 10, 2003 4:13 PM

» Normxxx - Re: Book: All About Market Timing

In response to message posted by radiodude:

That's right. Do the math. Let me know if the 200dma did better than buy and hold over the last 5 years. Don't tell me what it should have done! You remind me of those scientists that proved NO insects were capable of flight!

Besides, I happen to agree that trend following leads to all kinds of whipsaws and is not the best market timing technique. Which is why I use STS.

P.S. I have an M.A. in Mathematics and a Ph.D. in Mathematical Psychology.

And, oh yes, I taught statistics at the college level for two years.

-- posted by Normxxx

Top 292. Nov 10, 2003 4:35 PM

» Kirk - Stops and MAs are losing strategies

.
In response to message posted by Normxxx:

My belief is simple models that use 200 DMAs have worked well because we had a severe bear market.

I believe if we get a "trading range market" where we might run between 8,000 and 12,000 for the next 10 years, then any statistical market timing model will eventually slaughter folks that use it.

My guess is most will use 200 DMAs because they back test so well. Thus, the market will do its duty to suck the maximum number of dollars from those who think they can predict the future with trend following statistics and or find ways to save money.

As for circuit breakers... they can make you broke very, very fast. Just get a string of 5 bad markets in a row that just manage to trip your breakers and take you out at the limit before the market or security reverses.

As for leverage and options. Guess why I "trade" AMAT and LRCX?

I looked at index options which all the "experts" use. They have a time premium which eventually kills everyone.

Stocks are basically an option on the economy.

So stocks like Intel are basically options on the index funds because they have higher beta and folks buy PCs when expanding.

Now you might want to buy options on Intel for more leverage? That time penalty is there again. So, buy AMAT, LRCX, UTEK, etc. when low and take profits when high. IF you are wrong with a buy just before it cycles back down again, you as an individual investor can hold while mutual funds can't because they don't want to look like idiots. All I have to do is put up with morons calling me an idiot when my "option like stocks" are down and buy more. This place is good as the more I am called an idiot the more I buy.... Then when I am called a genius, I start to sell. It works pretty damned well. smile

I'd probably still be working for someone if I let stops (circuit breakers) take me out of my tech stocks. Remember mutual funds don't want to look stupid so they dump these shares when the market is against them... it hurts short term but it gives tremendous buying opportunities.

-- posted by Kirk