What is alpha? For those who spent their time in college perfecting the art of keg stands and discussing the finer points of beerpong, alpha may merely mean the first letter in the greek alphabet. For high priced lawyers and corporate rockstars, alpha provides a convenient description for their position at the top. However, to active investors and traders, alpha has special significance.
The alpha we are looking for is that extra return to be found in a market where inherent and predictable behavioral biases among both retail investors and fund managers result in mispriced assets. Classic economists and other academics would tell us that alpha is a myth. That the market is efficient and any additional returns we earn are due to luck or increased risk. I’d like to explore and challenge that assumption by discussing theories, strategies and by testing ideas both on paper and with my own money. Please join me in my quest to beat market averages over the long run.
Sunday, November 21, 2010
I did not expect the trends to hold, but all seven calls from last month would have turned out profitable. Every call and put I wrote expired worthless, and I would have netted quite a bit of money. I will try to post a second test tomorrow, as 7 trials does not a conclusion make.
Tuesday, October 26, 2010
Wednesday, October 20, 2010
I haven't successfully posted pictures in previous posts, but with my last post I committed. I just ran a test for some of my holdings using the 8 and 21 period EMA and the Heikin Ashi chart. Let's make a paper trade today based on real world prices.
Here is the 3 mo. chart for Silver Weaton (SLW). I've owned SLW since march 30, 2009: Now I have an 11% trailing stop order placed on the shares that aren't already covered by a call.
If we squint and look at the far end of the chart, we can see that we have two red bars, signaling a potential end to the 3 month upward trend (i had to chop a month off... sorry). In addition, the 8 period EMA has turned downward. This seems to indicate a potential trend reversal. At this point, I'm supposed to reduce my stop order to 4% from 11%. Ok. But for a test, we probably want something a little more exciting. Let's simulate a sale of calls, so there is some teeth in us being wrong in the direction. I'm going to simulate selling 3 contracts of SLW nov. 28 calls. currently selling for .86-.88. Since the stock stands at 26.68 right now, we have a little error on the upside. Selling 3 contracts nets me $253.74 (my brokerage charges me $5 per order, and .75 per option contract). For simplicity, we'll ignore profit potential in the next couple weeks, and look at the trade on expiration date (Nov 20).
Below is ARCC:
This time we see a continuation of close to a month long trend. Since we see a strong trend up, we'll go ahead and sell 3 contracts of puts this time.We'll sell 3 Nov 20 16.00 strike puts for .25 a piece, netting us $67.74.
Note: if you notice the highlighted spot, you can see a spot where an overeager trader may have exited his position and missed out on a run from 15- to the current price of 16.42.
Next: we'll test the australian ETF (EWA).
The next test will be SCCO.
ok. You know the drill. We have a trend up, so we'll sell puts. We can sell puts with an strike price of 41.00 at 1.70, netting us $502.74.
To save time I won't post DSX, QQQQ, DJX, but will track them as well. I realize 7 different test subjects do not a scientific study make. If this proves promising I will probably do a more extensive study come Nov. 20. Until then...
Anyways, I also planned on writing up a piece on momentum trading, and then testing trades in the same manner. While I was thinking about this, it occurred to me that my previous tests occurred during a very poor time to test just about ANY strategy against individual stocks.
See those tests I mentioned regarding Dr. Samir’s supposed tea leaf reading had, for the most part, predicted that chosen stocks would fall. I made paper option trades complying with these predictions (sold calls naked/built vertical spreads/ and sold puts). In the end, almost every stock rose instead, and the only trades Samir “predicted” accurately were those few that the method had just happened to predict would rise as well. Not surprisingly, as I started the trades when the market was falling, and I ended the trade in the middle of this recent run up.
As much as I like to hammer away at technical trading as “tea leaf reading”, I feel I have to be as fair as possible to Dr. Samir, and give his trades a fair shake; preferably at some time when individual stocks and the stock market aren’t so freakishly correlated (see last post).
Of course, no one is likely to write a follow up article stating “individual stock/ index correlation back to normal”, but I have noticed over the past week or two that the stocks in my portfolio have begun to swing a little more independently from the overall market. It’s not very scientific, but hey… if you have a better measurement, please send it my way. As such, this might be the ideal time to test a strategy sent to me by TD Ameritrade in their thinkMoney magazine.
I know what you are saying: “why trust a brokerage magazine… their goal is to get you to trade more, not to make more money”. That may be true, but they’ve included a trade that promises to look for the end of trends, and I’d like a shot at finding something to help me analyze trends, and evaluate another technical analysis “crystal ball” at the same time.
Trend analysis gets its start essentially from the practice of market timing. There are those of us who evaluate a stock and buy it when it seems undervalued, and those who try to buy a stock only when it is “about to go up”. Obviously, if you can time a stock with any reliability you would have a pretty big leg up in the game.
Efficient market proponents assure us that we can’t the market, but if you are still reading my posts, you know I don’t put a lot of stock in what they say. So let’s give the guys at TDAmeritrade a go:
Here’s the basic idea behind their technique. They use an 8 period and 21 period exponential moving average (EMA). This is basically the same as a moving average except that more weight is given to the more recent periods over older periods. In addition to the two EMA lines, this evaluation uses Heikin Ashi bars to display price action. Heikin Ashi bars were completely new to me, and I had to look them up at investopedia. These bars seem to be a form of averaging of very recent price movements incorporated into a candlestick chart.
Ameritrade depicts buying pressure candlestick bars as green and selling pressure candlestick bars as red. So, to find the end of a trend, Ameritrade tells us to look for two opposite colored bars against the current trend (if a trend is down, we are looking for two green… if a trend is up, we are looking for two red). Once we have found two of the opposing bars, we are to look at the 8 and 21 period EMA. If the 8 period EMA is changing directions, than we are to move our stops within 4% of the current price. When 8 period EMA crosses the 21 period EMA, we are to consider the trend over and pull our position, if it hasn’t been stopped out already.
Basically, it looks like this method is saying “If Stock was going up, but now it’s going down… and it’s going down more now than it was going down yesterday, you might want to sell it.” Got it. We’ll give it a shot.
Sunday, October 3, 2010
argues that stockpickers are having increasing difficulties in beating market averages because stocks are now moving in correlation with the overall market at a very high level. In fact, the article states that correlation between individual stocks and the market have reached a peak of 74% within the past few months, compared to an average of 27% between 2000 and 2006.
I must confess that I cannot be categorized as a “buy and hold” investor. I try to hold onto my investments as long as it makes sense, but as soon as valuations reach the point where an alternate investment is likely to yield higher returns, I dump my investment and switch.
However, I think this article heralds a golden opportunity for long term buy and hold investors, despite James Bianco’s assertion that “stock picking is a dead art form”. My logic is simple:
We should chose stocks based on price compared to free cash flow. If a company’s free cash flow is growing at a greater rate than the stock market is pricing it up, than the company is slowly being undervalued. In this environment, where investors are ignoring stock fundamentals, than companies with low debt, who are successfully grabbing market share from failing competitors should become increasingly underpriced. This trend could go on for years, but the longer it goes, the greater the opportunities. Eventually, the mispricing will become attractive enough to investors to shift money out of their holdings in precious metals and bonds. Than the long term investor’s positions pay off.
The risks are obvious:
1. The correction takes years, and the company’s competitive advantage does not hold up, yielding the long term investor anemic returns when considered over the time period.
2. The investor picks the wrong stocks, mistaking financial strength for accounting shenanigans.
3. The trend continues and investment valuations are “changed” forever
Both risk 1 and 2 exist in any circumstance, and I believe careful stock screening and picking will minimize these risks. I’ve laid out my stock screen in the past post, and I think that it goes a long way in identifying survivable stocks that will pay you while you wait for prices to equalize. In addition, you will need to monitor the health of your holdings frequently to ensure you’re not holding a dud in the making.
As for risk 3, I don’t believe for a second that investor methodology has changed for good. In the long run, no one will pay less for a cash flow that is increasing at a greater rate than its peers and has more certainty of continuing to do so. Over the past 15 years, many people have asserted that the laws of supply and demand, and the laws of mathematics are dead. They have all proven wrong and have paid the price.
Here’s a link to a video on the same topic: (http://online.wsj.com/video/is-stock-picking-a-dead-art-form/7551169A-451A-4FA2-AE29-19E645303FCB.html)
Tuesday, September 28, 2010
First, I use a screener to find potential investments. I have used several in the past, but right now I prefer the one found at www.finviz.com.
Finviz divides its screens into 3 broad categories. They are descriptive, fundamental, and technical. Before we begin the screening, we can see a total of 6730 stocks available to us.
I start with descriptive screens. I tend to leave all options blank except for the following: Dividend yield, Average volume, and option/short.
I start by setting dividend yield to >0. I prefer to have a stock pay some sort of dividend even though theoretically and logically, dividend payments should not make any difference to an investor. Only free cash flows should matter, not the manner they are paid out.
Next, I set average volume to Over 1M. 1 million shares trades on average is not extremely fluid, but I don’t want to limit our pool to only the most traded stocks either. Still, choosing stocks with low volume has its costs. Bid/ask spreads are bound to be wider, and we are likely to see surprise jumps in prices over what we expect to pay with market orders. This extra cost is called slippage, and is the reason I tend to make limit orders only. I lose opportunities from time to time, but I find limit orders to be well worth the occasional loss of a trade.
Finally, I set the option/short bar to option: This limits the presented stocks to those with option contracts available. I like to be able to sell covered calls, as well as hedge positions on occasion.
I understand that these three screens might be somewhat controversial. If you don’t care about dividends or options, there is definitely no problem with removing them from your screen. Using them in your screen no doubt removes some great potential trades/investments, but they are integral to my trading style, so I’ve included them. As for volume, you can determine what you consider a tolerable average, but I think you should have some kind of cap.
With these screens in place, we’ve limited our selection to 628 stocks (see below).
I next enter the fundamental screens. These screens are the heart of finding a good “underpriced” stock.
There are three aspects I am going to screen stocks for: price, profitability, and solvency.
To screen for price (we want a low price to performance ratio) I use 2 basic screens.
The first screen is the infamous P/E ratio. I set P/E ratio to 15 or less. I would have to have 12 or less, but this is not an option. Since the average P/E ratio for the S&P 500 over the last 130 years has been 12 (according to Jeremy Segal in “Stocks for the Long Run”), I would prefer to find stocks with P/E ratios of 12 or less.
The next screen I use is Price/FCF. I set Price/FCF to 15 or less. Since rational investors should base investment valuations on free cash flow, it only makes sense that free cash flow should be taken into account somewhere in our screener. Besides, the Earnings (the e in p/e), can be manipulated through overleveraging or accounting shenanigans. The lower the number in this case, the less you are paying per $1 of free cash flow generated per share.
Next, we set out to measure profitability. I like to measure profitability of a company by measuring Economic Value added (which is a company’s profit minus its costs of capital). Such a measurement is very hard to manipulate. Unfortunately, EVA is not an option on any screener I can find, so we will need to settle for return on equity.
I set my return on equity screen to over 10%. 10% return is my required rate of return, so any company I invest in had better be able to return more than that on their owner’s stake. Any earnings reinvested into the company should be grown at this rate, which is why ROE is important. Unfortunately, it can be distorted or manipulated through share repurchases, large writeoffs, or by increasing leverage. Therefore, take high ROE numbers with a grain of salt.
Finally, I prefer to own stock in a company that is not flirting with bankruptcy. Therefore, I have three tests for a company’s solvency.
First, I prefer a company with a debt to equity ratio under 40%. However, if the company is performing according to the prior screens, I can tolerate a little more debt.
So I input a screen of debt/equity Under 0.5. While debt provides tax benefits because a company’s interest is deductable, too much debt raises costs of debt and limits a company’s options during hard times.
In addition, I run another screen to eliminate any companies that are experiencing cash flow issues. By testing a company’s current ratio. The current ratio (current assets/current liabilities) tests a company’s abilities to pay short term debt with liquid assets. Anything over 1 should be healthy.
I set my screen up as Current Ratio: Over 1.5
A final, if redundant test is the quick ratio. Quick ratio is calculated as Current Assets - inventories/current liabilities. Quick ratio also measures a company’s ability to pay back liabilities with its most liquid assets. Because the quick ratio takes into account inventory, it may be a more appropriate measurement of liquidity than the current ratio for some business models.
I set Quick Ratio: Over 1
With these inputs, we’ve narrowed our pool down to 20 stocks. Unfortunately, I’ve found no technical screens that seem remotely reliable. So we now must sort through our 20.
Tuesday, September 14, 2010
Something smelled fishy to me on that day. And the more I learned about the efficient market theory, the more convinced I was that the market could be beaten. While the successes of Warren Buffett, Jim Rogers, George Soros, and Peter Lynch could be the result of extremely long bouts of luck, I just didn’t believe it. I had a gut feeling that people were not behaving rationally, and that there was money to be had in exploiting that.
The belief that investors can’t beat the market without taking on additional risk stems from the “efficient market hypothesis” popularized by a University of Chicago professor, Eugene Fama in the 1960’s. Efficient market theory states that, since all information needed to price equities is equally available to all participants, all information will be priced into stocks instantaneously. Any illogical mispricing that develops will be exploited by other rational investors (arbitrage), until the mispricing corrects itself. Not all investors need be perfectly logical; all that is needed is that irrational actions within the market should be random, and rational investors will not be able to predict aberrations in price in the long run.
The hypothesis makes sense to me only if the following hold true:
A. Most, if not all irrational behavior is random and unpredictable
B. Arbitrage is possible. That is, exploiting obvious mispricing is riskless
C. Information is reliable, timely, and available to everyone on equal terms
Obviously, I don’t believe that these three conditions exist. First, I recognized early on that people did not behave in the rational manners that economic models relied on. I recognized overconfidence in investors early on, as I begin investing in the runup to the tech bubble burst in 2000. Warren Buffett had written a description of both market bubbles and market cycles in a case study we were required to read, and I spent months arguing to friends that the market had to crash at some point. I was meant with anger arguments, from poly-sci majors with a passion that exceeded even religious debates. If ordinary people could vehemently oppose my suggestion that 20% growth could not continue indefinitely, than I knew that something truly powerful was pushing markets beyond reason.
As for the second point, I don’t want to delve too deeply in the topic yet. I feel that arbitrage deserves its own post, and I’ll try to get to it as soon as possible. Let me just give a related, if simplistic and perhaps inaccurate example. As discussed above, I was under the impression that a crash was imminent. The problem was, I didn’t know when. I’d read claims from 1997 that stocks were overpriced, and I’d read the scathing critiques against those “perma bears”. At the time, if one wanted to bet against the stock market, one had three choices.
A. Short several stock or the index
B. Purchase calls on stocks or the index
C. Get out of the market
As a young, college student, I wasn’t ready to sell stocks short or purchase puts. Besides, these routes require you to be pretty dead on timing wise. If shorting a stock, one faces interest rates and margin calls. The position is not tenable forever. If purchasing puts, the time limit is more obvious. Turn a profit prior to the exercise date; or else.
I didn’t have money in the market, so I chose to stay out of the market. It paid off, but it isn’t a strategy that generates much money. Besides, had my timing been off for a few more years, it might have proved a costly strategy as well.
The third point is that information must be available to all market participants at the same time, and that it must be timely and reliable. Again, this is a topic best discussed by itself, but let’s examine it briefly, none the less. Do we really believe that no information makes it to some portion of investors prior to its public release? Do we really believe that each piece of information can be properly digested and given proper weight amongst the thousands of other pieces of chaff and noise information spewed by the 24 hour news machine that is cable and the internet? Do we believe that complex industry information can be equally analyzed by all investors in that particular industry, or do we recognize that certain parties will be able to interpret the information far more efficiently than the vast majority of participants?
If we do not believe that the efficient market hypothesis holds, than there may be room to exploit inefficiencies in the market. We can’t guarantee success, but we should be able to reduce risk inherent in a portfolio, while increasing returns. But we will need to understand how people react to specific situations, so that seemingly random irrational behavior may become somewhat more predictable.
Fortunately, behavioral finance, explored by Amos Tversky and Daniel Kahneman in prospect theory, offers a glimpse into the human decision making process and how it applies to decisions made regarding money. If we examine the findings in prospect theory and other behavioral studies, we may be able to find trades that hold higher expected returns (total payoff X probability).
I would be interested to hear any thoughts on this matter, especially defenses of the efficient market hypothesis.