Market Timing: Should You Attempt It?
Read the article this discussion is about
- Normxxx
- radiodude
- Normxxx
- Kirk
- radiodude
- Kirk
- Normxxx
- Kirk
- Normxxx
- Normxxx
This archived discussion is "read only".
For the corresponding "live" discussions, post in the active topic forum here.
« Previous 23 24 25 26 27 28 29 30 31 32 33 34 35 Next »
» 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
» 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
» 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
» 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. 
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
» radiodude - Re: Re: Book: All About Market Timing
In response to message posted by Normxxx:Good, I think we can take the converstaion to a little more depth.
I'm sure you are familiar with the phase delay associated with a Linear Phase finite impulse response discrete low pass filter. In this case, you have a 200 tap discrete filter with equal weighting on each tap. As you know, the phase delay is quite large for this arrangement, so I don't see how this output is useful to anyone.
Sure, 200dma have worked over the last 5 years, but again because of the phase delay, you would have moved out after a movement downward, and you would have gotten back in after a rise.
The tools that worked in the past 5 years may not be applicable to the future.
Also, for the last 5 years, well, CD's did better, so should we never invest in stocks based on this data?
I'm not saying what any model should have done, rather I just don't want you to be fooled by randomness and/or data mining.
-- posted by radiodude
» Kirk - Re: Re: Book: All About Market Timing
In response to message posted by Normxxx:That's right. Do the math. Let me know if the 200dma did better than buy and hold over the last 5 years.
Out of curiosity, how have the "Dogs of the DOW" done the last 5 and 10 years?
I remember that was all the rage back in 1998 when I left HP. I think dividend stocks really sucked between 1998 and 2000. They sucked just long enough for folks to give up on that DOG idea back tested model before the bear hit. I wonder if it is ahead now?
-- posted by Kirk
» Normxxx - Re: Stops and MAs are losing strategies
In response to message posted by Kirk:Well, you are not going to get me to defend SMAs. All I am saying, is that over the long haul (and not just the last 5 years) they work better than buy and hold. As for the hypothetical market you assumed, if you can assume a market that will kill the the smas, I can assume a market that will kill your approach (assume that the market P/Es drop 75% or so and don't come back). In fact, I don't have to assume, I can just point to the Nikkei!
Unless, of course, you do judicious stock picking (and don't ride them to hell if you think they are not coming back). But a lot of less than effective stock pickers saw their stocks lose 90-95% of their values, and are still down a ton.
Since I didn't have the time to stock pick and I don't trust anyone else to do it for me, I generally trade the indexes (only occasionally the options, and then I prefer to sell them). Maybe I will subscribe to your newsletter. If JJ likes it, it must really be something!
As you know I use the Seasonal Timing System (for almost the last 4 years), which has sailed through this Bear brilliantly. None of my circuit breakers were triggered in the last 10 years or so (about the time I have been using them), but largely because I was already out of the market when they would have triggered. Yes, I am well aware of the pitfalls of stop losses and trend following models, which is why I do not use the one and make judicious use of the other.
My circuit breakers are not stop losses; they use ma's, but I use a set of them (4 to be exact), and they are a last resort (so I can set a really low threshold). (I swear I will do a writeup of them within the next couple of weeks.)
STS works fine in a "trading range market." It has proven itself in all kinds of markets over the last 50 years.
You're right about stocks being an option on the economy, but there is nothing to prevent the P/E ratios from falling 75% or so and not coming back. Stocks have no intrinsic value. Unless you can find a buyer, you are stuck with them. (See my post on how stocks became a giant Ponzi scheme.)
I know a lot of people that have espoused your philosophy of buy and hold and are currently still down 75% or so. I am related to one of them. Does anyone know where the average fund is today compared with January 2000?.
I don't know why you insist that what you are doing is "buy and hold." It seems to be anything but. You buy when they are low, you sell when they are high, maybe buying some back on a pullback if you still like them. Granted, your selling strategy (assuming you still like them, unlike LU and T) is more of a money management strategy, but it is NOT "buy and hold," which takes no heed of any other strategy. I.e., as interpreted by Wall Street, "buy and hold" is really "buy and hold forever!" Haven't you read the cautionary tales where little old ladies died as multimillionaires because they refused ever to sell their stock!
-- posted by Normxxx
» Kirk - Re: Re: Stops and MAs are losing strategies
In response to message posted by Normxxx:I don't think I have ever claimed I was buy and hold.
What I recommend for others is an asset allocation strategy where you buy and hold BUT rebalance between stocks and bonds when one goes on a run driving your target allocation out of whack.
Have you read these?
What I do takes a great deal of time. The returns for me have been better than buying and holding while working a job so I continue. The newsletter helps me articulate my thoughts and has done pretty well. Some have even found value in it.
Many read it and still don't get the concept of a total portfolio and risk management so they don't do so well as they are looking for a list of stock picks.
-- posted by Kirk
» Normxxx - Re: Re: Re: Book: All About Market Timing
In response to message posted by Kirk:Actually, the last I saw (about 2 years ago) it was working again after NOT working for a couple of years. (It did not fail by much though, as I recall.) But the DoD is measured against the Dow, so you can have down years even when it is working. I don't think I like that. (Yes, ma'am, the operation was a success but the patient died.) Anyway, the DoD approach is only spectacular when measured against the average fund, which is usually beaten by all of the averages.
-- posted by Normxxx
» Normxxx - Re: Re: Re: Book: All About Market Timing
In response to message posted by radiodude:Please re-read my earlier posts. I do not, repeat do not use a SMA for timing. That was strictly a worst case example to show that B&H sucks. I suggest you thank God every night that you are not Japanese.
-- posted by Normxxx
« Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Next »
Please follow the guidelines set forth in the Suite101 Posting Etiquette when adding to the discussion.