A Failed Model And A Cautionary Tale

Includes: DIA
by: Jamin Chen

In a comment to my recent article, a reader said I was emotional for not participating in the current upswing in the market. The tale below serves as a response to him, as well as a cautionary tale, at the least, for me.

Testing a model to predict the market movement is like performing a physics experiment. However, there is a huge difference. In physics, once a principle is established, its prediction holds true all the time. In market behavior, it may hold true but only sporadically.

Here is my tale of the testing of a market model.

While QE2 was ongoing, I hit upon an idea that the money the Fed had been injecting into the economy was really pumping liquidity into the market. Therefore, there should be a correlation between how much the Fed injected and the valuation of the market. I, therefore, faithfully added up all the money that the Fed had injected and plotted it against the market index, DJIA. I did this first thing in the morning every day. On February 8, 2011, the chart looked like this:

A good correlation, I told myself. Convinced, as QE2 was to run through the end of June 2011, I planned my investment accordingly. I sold off the entire DOG I then owned for a loss and bought a bunch of DIA.

What happened after that was this:

In fact, I kept buying DIA until it hit a high of $128. Right after that, the market turned definitely south. DJIA first dipped to about 11,500 at the end of QE2 (June 2011). It then dipped further to about 10,750 in August-October, 2011. See the chart below from Yahoo!Finance.

Then, in September 2011, the Fed launched Operation Twist. It was supposed to run through June of this year but it got extended. Ever since it was launched, the market has been zooming up and it closed at 12,950.10 on February 17 as shown above. It is an election year. Some say the market goes up on an election year. The market response to Operation Twist so far is very much like the first half of QE2. Would the market response to the rest of Operation Twist be like that of QE2?

Would I wholeheartedly dive into the market at this time? Would you? I would need my head examined if I do. I would rather miss it if the market does keep going up. I don't want to get burnt twice.

The model failed. It failed not because its underlying principle was incorrect. It failed because the market does not move by one parameter alone. It is a lot more complex. It is affected by many parameters. Only that, once in a while a specific parameter may have an overwhelming influence upon it and after a while other parameters take over. I need new plausible models to look into where the market is going next. Until then, I should sit tight and just watch the market moves on.

My current view of the market behavior is like this. When there is a seller there is a buyer. So, who were the buyers in August-October of 2011 low? Who were the buyers back in June-September 2010 low? And, who were the sellers in April-July 2011 high? I bet there are some really smart people who did just that. I want to follow what they did.

Looking back, I should have accumulated tons of DIA in August-September of 2011 low and sit tight and wait for the next top to unload them. But I am just an infinitesimally small fish in the pond. I cannot move the market. Also there is no way I can call the top or the bottom. Nor I would know when the top is or when the bottom is even when I am in the midst of it. All I can do is follow the market and roll with the punches.

So, looking at the chart above, I tell myself now may be about the time when I should start unloading my DIA, had I accumulated enough of them before. Since I hadn't, I am very cautious about dipping my toe into this rising market. I first have to come up with a darn good strategy to take advantage of the remainder of this bull market. The strategy must have clearly defined entry and exit points and it must also limit the exposure when the bull market ends. By exposure I mean the shares I may buy at the top of the market. My previous article is just one such attempt.

Before I dip into the market, I have to figure out what kind of risk I will be taking. Even though I will be buying on the dip, I need to set the highest price I would buy. For example, I would buy at DIA as high as $137 if DIA goes up to $140 and then drops back to $137. I would stop buying even if the market keeps moving up from there. Next, I need to set the lowest price I would buy. For example, I would buy at DIA as low as $100. Below that, I would also stop buying. Between the top and the bottom, suppose I would buy 100 shares of DIA for every drop of $3, what kind of risk would I be taking?

The worst scenario with this strategy is something like this: If the market goes straight up to DIA=$140 and then tumbles straight down to DIA=$100, the most I would buy is about 1,200 shares of DIA at an average price of about $120. There would be a paper loss of about $24,000 when DIA hits $100. This is not an inconsiderable amount of loss. However, judging from the behavior of the market in the recent past, though not a sure thing, I may be able to recover most of that loss in about 6 months after the crash. That is the risk I would be taking. This risk could be real, just look at the way the market tumbled roughly from DIA=$128 to $107 in August 2011, as shown in the chart above. Is this risk worth the reward?

The main reward of this strategy is when the market zigzags between the top and the bottom I chose, instead of moving straight up and down.

I may choose a different set of top and bottom with a different possible exposure I may have. If I choose a range that is very narrow, I may miss the next zigzags of the market.

It is important to know that, by setting up a strategy like this, it does not mean the market is going to behave as expected. However, by having a strategy, you know what your possible maximum exposure is and the possible market zigzags that you may be able to catch.

There are many variations to this strategy as well as other strategies to deal with the current rising market. For example, instead of buying on the dip, you may simply buy 100 shares of DIA for each $3.00 rise in DIA from here on and place an ever higher stop loss sell order as the market rises. This could be a better strategy if the market does go straight up and down in the next few months. Even with this scheme, when the market tumbles to a certain level, you may go back to the buying on the dip strategy. Again, if you were to adopt any of these strategies, you will have to go through all the possibilities you can dream up and weigh the risk of each of them and make sure the risk is worth taking.

Disclosure: I am long DIA.