Some Elaborate Technical

Some More Elaborate Technical Systems

Since many of the technical analysis systems tested are very popular, let’s briefly examine a few in detail.

The Filter System

Under the popular “filter” system a stock that has reached a low point and has moved up, say 5 percent (or any other percent you wish to name here and throughout this discussion), is said to be in an uptrend. A stock that has reached a peak and has moved down 5 percent is said to be in a downtrend. You’re supposed to buy any stock that has moved up 5 percent from its low and hold it until the price moves down 5 percent from a subsueqent high, at which time you sell the stock and, perhaps, even sell short. The short position is maintained until the price rises at least 5 percent from a subsequent low.

This scheme is very popular with brokers, and forms of it have been recommended in a variety of investment books. Indeed, the filter method is what lies behind the popular ll “stop-loss” order favored by brokers, where the client is advised to sell his stock if it falls 5 percent below his purchase price to “limit his potential losses.” The argument is that presumably a stock that falls by 5 percent will be going into a downtrend anyway.

Exhaustive testing of various filter rules based on past price changes has been undertaken. The percentage drop of rise that filters out buy and sell candidates has been allowed to vary from 1 percent to 50 percent. The tests covered different time periods from 1897 to the present, and involved individual stocks as well as assorted stock averages. Again, the results are remarkably consistent. When the higher brokerage commissions incurred under the filter rules are taken into consideration, these techniques cannot consistently beat a policy of simply buying the individual stock (or the stock average in question) and holding it over the period during which the test is performed. The individual investor would do well to avoid employing any filter rule and, I might add, any broker who recommends it.

Elaborate System. Photo by Elena

The Dow Theory

The Dow theory is a great tug-of-war between resistance and support. When the market tops out and moves down, that previous peak defines a resistance area, since people who missed selling at the top will be anxious to do so if given another opportunity. If the market then rises again and nears the previous peak, it is said to be “testing” the resistance area. If, on the other hand, the market “fails to penetrate the resistance area”, and instead falls through the preceding low where there was previous support, a bear-market signal is given and the investor is advised to sell.

The basic Dow principle implies a strategy of buying when the market goes higher than the last peak and selling when it sinks through the preceding valley. There are various wrinkles to the theory, such as penetration of a double or triple top being especially bullish, but the basic idea is followed by many chartists and is part of the gospel of charting.

Unhappily, the signals generated by the Dow mechanism have no significance for predicting future price movements. The market’s performance after sell signals is no different from its performance after buy signals. Relative to simply buying and holding the representative list of stocks in the market averages, the Dow follower actually comes out a little behind, since the strategy entails a number of extra brokerage costs as the investor buys and sells when the strategy decrees.

The Relative-Strength System

Here an investor buys and holds those stocks that are acting well, that is, outperforming the general market indices in the recent past. Conversely, the stocks tat are acting poorly relative to the market should be avoided or, perhaps, even sold short. While there do seem to be some time periods when a relative-strength strategy would have outperformed a buy-and-hold strategy, there is no evidence that it can do so consistently. A computer test of relative-strength rules over a twenty-five-year period suggests that such rules do not, after accounting for brokerage charges, outperform the placebo of a buy-and-hold investment strategy.

Price-Volume Systems

These strategies suggest that when a stock (or the general market) rises on large or increasing volume, there is an unsatisfied excess of buying interest and the stock can be expected to continue its rise. Conversely, when a stock drops on large volume, selling pressure is indicated and a sell signal is given.

Again, the investor following such a system is likely to be disappointed in the results.The buy and sell signals generated by the strategy contain no information useful for predicting future price movements. As with all technical strategies, however, the investor is obliged to do a great deal of in-and-out trading, and thus his brokerage costs are far in excess of those necessitated in a buy-and-hold strategy. After accounting for these brokerage charges, the investor does worse than he would by simply buying and holding a diversified group of stocks.

• Burton G. Malkiel. A Random Walk Down Wall Street, including a life-cycle guide to personal investing. First edition, 1973, by W.W. Norton and company, Inc

Are Security Analysts Clairvoyant?

Are Security Analysts Fundamentally Clairvoyant?

Forecasting future earning is the security analysts' raison d'être. As a top Wall Street professional put it in his fraternity magazine, Institutional Investor: “Expectation of future earnings is still the most important single factor affecting stock prices.” Growth in earnings and therefore in the ability to pay dividends is the key element needed to estimate a stock's firm foundation of value. The analyst who can make accurate forecasts of the future will be richly rewarded. “If he is wrong,” Institutional Investor puts it,” a stock can act precipitously, as has been demonstrated time and time again. Earnings are the name of the game and always will be.”

To predict future directions, analysts generally start by looking at past wanderings. “A proven score of past performance in earning growths is,” one analyst told me, “a most reliable indicator of future earnings growth.” If management is really skillful, there is no reason to think it will lose its Midas touch in the future. If the same adroit management team remains at the helm, the course of future earnings growth should continue as it has in the past, or so the argument goes.

Such thinking flunks in the academic world. Calculations of past earnings growth are no help in predicting future growth. If you had known the growth rates of all companies during, say, the 1970-80 period, this would not have helped you at all in predicting what growth they would achieve in the 1980-90 period. And knowing the fast growers of the nineties. This startling result was first reported by British researchers for companies in the United Kingdom in an article charmingly titled “Higgledy Piggledy Growth.” Learned academicians at Princeton and Harvard applied the British study to U.S. Companies – and, surprise, the same was true here!

An old bank. Photo by Elena

“IBM,” the cry immediately went up. “Remember IBM!” I do remember IBM: a steady high grower for decades. For a while it was a glaring exception. But after the mid-1980s even the mighty IBM failed to continue its dependable growth pattern. I also remember Polaroid and dozens of other firms that chalked up consistent large growth rates until the roof fell in. I hope you remember not the exception but rather the rule: There is no reliable pattern that can be discerned from past records to aid the analyst in predicting future growth.

A good analyst will argue, however, that there's much more to predicting than just examining the past record. Rather than measure every factor that goes into the actual forecasting process, John Cragg and I decided to concentrate on the end result: the prediction itself.

Donning our cloak of academic detachment, we wrote to nineteen major Wall Street firms engaged in fundamental analysis. The nineteen firms, which asked to remain anonymous, included some of the major brokerage firms, and banks engaged in trust management. They are among the most respected names in the investment business.

We requested – and received – past earnings predictions on how these firms felt earnings for specific companies would behave over both a one-year and a five-year period. These estimates, made at several different times, were then compared with actual results to see how well the analysts forecast short-run and long-run earnings changes. The results were surprising. Bluntly stated, the careful estimates of security analysts (based on industry studies, plant visits, etc.) do very little better than those that would be obtained by simple extrapolation of past trends, which we have already seen are no help at all. Indeed, when compared with actual earnings growth rates, the five-year estimates of security analysts were actually worse than the predictions from several naïve forecasting models.

For example, one placebo with which the analysts' estimates were compared was the assumption that every company in the economy would enjoy a growth in earnings approximating the long-run rate of growth of the national income. It often turned out that if you used that naïve forecasting model you would make smaller errors in forecasting long-run earnings growth than by using the professional forecasts of the analysts.

Our method of determining the efficacy is exactly the same as was used before in evaluating the technicians' medicine. We compared the results obtained by following the experts with the results from some naïve mechanism involving no expertise at all. Sometimes these naïve predictors work very well. For example, if you want to forecast the weather tomorrow you will do a pretty good job by predicting that it will be exactly the same as today. It turns out that while this system misses every one of the turning points in the whether, for most days it is quite reliable. How many weather forecasters do you suppose do any better?

When confronted with the poor record of their five-year growth estimates, the security analysts honestly, if sheepishly, admitted that five years ahead is really too far in advance to make reliable projections. They protested that while long-term projections are admittedly important, they really ought to be judged on their ability to project earnings changes one year ahead.

Believe it or not, it turned out that their one-year forecasts were even worse than their five-year projections. It was actually harder for them to forecast one year ahead than to estimate long-run changes.

The analysts fought back gamely. They complained that it was unfair to judge their performances on a wide cross section on industries, since earnings for electronic firms and various cyclical companies are notoriously hard to forecast. “Try us on utilities, one analyst confidently asserted. So we tried it, and they didn't like it. Even the forecasts for the stable utilities were far off the mark. Those the analysts confidently touted as high growers turned out to perform much the same as the utilities for which only low or moderate growth was predicted. This led to the second major finding of our study: There is not one industry that is easy to predict.

Moreover, no analysts proved consistently superior to the others. Of course, in each year some analysts did much better than average, but there was no consistency in their pattern of performance. Analysts who did better than average one year were no more likely than the others to make superior forecasts in the next year.

My findings with Cragg have been confirmed by several other researchers. For example, Michael Sandaretto of Harvard and Sudhir Milkrishnamurthi of M.I.T. Completed a massive study of the one-year forecasts of the most widely followed companies between 1977 and 1981. The number of companies monitored was about 1,000 each year, and, in general, estimates were available from five or six analysts for each company. All estimates were made for the then current year, so that 1981 estimates had been made early in 1981. The staggering conclusion of the study was that the average annual error of the analysts was 31.3 percent over the five-year period. The error rates each year were remarkably consistent – the lowest error rate was 27.6 percent in 1978, the highest 33.5 percent in 1981. Financial forecasting appears to be a science that makes astrology look respectable.

Amidst all these accusations and counter-assertions, there is a deadly serious message. It is this: Security analysts have enormous difficulty in performing their basic function of forecasting earnings prospects for the companies they follow. Investors who put blind faith in such forecasts in making their investment selections are in for some rude disappointments.

Burton G. Malkiel. A Random Walk Down Wall Street, including a life-cycle guide to personal investing. First edition, 1973, by W.W. Norton and company, Inc.

Preeclampsia

Preeclampsia

Another Reason to Make Love

Your partner’s sperm may protect you against severe complications

Even women with no history of hypertension can find the physical and emotional demands of pregnancy to be a recipe for high blood pressure. Untreated, this hypertension can lead to severe complications for both the mother and fetus. And in roughly 1 in 10 pregnancies, a dangerous condition called preeclampsia can develop during the last trimester.

Symptoms include not only hypertension, but blurred vision, swelling in the face and hands as a result of fluid retention, and protein in the urine. Headaches, nausea, vomiting, and abdominal pains may also arise, and, if neglected, this toxemia can lead to full-blown eclampsia, a condition that can produce seizures that are potentially fatal to both the mother and the child.

Now a study published in a respected medical journal, the Lancet had put forth a possible explanation for preeclampsia. Following up on earlier research suggesting that the condition occurs more frequently in women who have never before been pregnant or are pregnant by a new partner if the have been pregnant previously, researchers studied a group of women in Guadeloupe, where it is not unusual for women to have children by many partners. Because of the prevailing social attitude there about having children with multiple partners, the investigators were able to gather unusually detailed information about their subjects’ sexual history for the study.

Goddess of Love. Photo by Elena

Among women pregnant for the first time, the authors of the Guadeloupe study found preeclampsia 12 percent of the time, compared with only five percent for those who were pregnant by the same man for a second time. Even more striking, the researchers discovered preeclampsia in 24 percent of the women who had had children before, but were pregnant this time by a new partner.

Delving further, the scientists found that the longer a woman had been sexually intimate with the man who had made her pregnant, the less her risk of preeclampsia.

While little is known about the biological mechanisms that trigger preeclampsia, scientists have established that the placenta fails to attach properly to the uterine lining in such instances. The authors of a study, led by French physician Pierre-Yves Robillard, suspect that preeclampsia may be an adverse immunological reaction by the mother against certain genetic material implanted in the embryo by her partner’s sperm.

This could explain why the placenta does not fully attach to the uterine lining in cases of preeclampsia, resulting in heightened blood pressure in the woman’s late in pregnancy, as her body seeks a way to supply nourishment to the fetus. Based on the Guadeloupe data, the Robillard team hypothesizes that the more exposure a woman has to a partner’s semen before becoming pregnant, the more likely she is to develop an immunity to the man’s genes, thereby reducing her chances of developing preeclampsia.

Not everyone in the medical community agrees with the scientific hunch. Some researchers suggest that the condition is a largely inherited disease, and others say that it may be a response to a lack of protein or too much salt in the mother’s diet. But in any case, the Guadelope study offers intriguing support for the maxim “You dance with the one who brung you.”

A lady in love. Photo by Elena

Why Might Charting Fail to Work

Why Might Charting Fail to Work?

It is easier for me to present the logical arguments against charting. First, it should be noted that the chartist buys in only after price trends have been established, and sells only after they have been broken. Since sharp reversals in the market may occur quite suddenly, the chartist will miss the boat. By the time an uptrend is signaled it may already have taken place. Second, such techniques must ultimately be self-defeating. As more and more people use it, the value of any technique depreciates. No buy or sell signal can be worthwhile if everyone tries to act on it simultaneously.

Moreover, traders will tend to anticipate technical signals. If they see a price about to break through a resistance area, they will tend to buy before, not after, it breaks through. If it ever was profitable to use such charting techniques, it will now be possible only for those who anticipate the signals it is doubtful that any profitable technical trading rules can be developed.

Charting may fail to work. Photo by Elena

Perhaps the most telling argument against technical methods comes from the logical implications of profit-maximizing behavior on the part of investors. Suppose, for example, that Universal Polymers is selling at around 20 when Sam, the chief research chemist, discovers a new production technique that promises to double the company's earnings and stock price. Now Sam is convinced that the price of Universal will hit 40 when the news of his discovery comes out. Since any purchases below 40 will provide a swift profit, he may well buy up all the stock he can until the price hits 40, a process that could take no longer than a few minutes.

Even if Sam doesn't have enough money to drive up the price himself, surely his friends and the financial institutions do have the funds to move the price so rapidly that no chartist could get into the act before the whole play is gone. The point is that the market may well be a most efficient mechanism. If some people know that the price will go to 40 tomorrow, it will go to 40 today. Of course if Sam makes a public announcement of his discovery as the law requires, the argument holds with even greater force. Prices may adjust so quickly to new information as to make the whole process of technical analysis a futile exercise.

Illustration: Megan Jorgensen

Placebo Effect

Placebo Effect

The word placebo has many meanings and a rich history. It originates in a tale of mistranslation, religion and behavioural idiosyncrasies. Placebo controls may have helped to dehypnotize the public after mesmerism (Franz Anton Mesmer) and animal magnetism ‘medical’ sensations in the late 18th century (Finniss et al., 2010). A resembling term, nocebo, stands for an adverse or unwanted response to pharmacological or therapeutic treatment.

The medical definition implies a sugar pill or something else non-medicinal, that is supposed to be the medication. Placebos are often used in clinical studies to distinguish true effects of treatment from the comfort, expectancies, beliefs and perceptions that the subject may have. A limitation of such methodology is that the glucose level change affecting cerebral metabolism may confound the variables.

Placebo effects can be true and perceived. A self-fulfilling prophecy may also be involved. A patient may suffer from the flu and take a cold and sinus pill. Behind the scenes, unknown to this person, he or she may actually be sick because of stress. For optimal immune system function, psychological stability is important too. Reassured by the substance, the individual feels better in both senses.

Gothic Sphere. Placebo Effect. Photo: Elena

Mayberg et al. (2002) conducted Positron Emission Tomography (PET) scans to elucidate which brain areas show the highest activity during administration of placebos. Cortical areas that were most responsive: prefrontal, anterior cingulate, premotor, parietal, posterior insula and posterior cingulate. Least responsive: subgenual cingulate, parahippocampus and thalamus (‘the brain’s relay station’).

PET is a neuroimaging technique widely used in neuroscience along with functional Magnetic Resonance Imagery and (fMRI) and Computerized Axial Tomography (CAT). Some of the machinery takes up a whole room, but the sliced photographs greatly contribute to health care and scientific discovery.

Moerman & Jonas (2002) narrate that the meanings ascribed to the word pair are highly misguided. The duo proposes a new formulation “meaning response”. The placebo may be therapeutic in itself, if that (ever) is the right prescription.

In their review paper, Finniss et al. (2010) make it clear that the preconceptions persons hold can both mediate and modulate placebo effects. Classical conditioning is also listed as associated with placebo mechanism.

Classical conditioning dates back to the Russian Empire, where Ivan Pavlov trained dogs to salivate to the sound of a bell (conditioned response), because they were conditioned to associate the sound of the bell (conditioned stimulus) with food (unconditioned stimulus). In canines, salivation in anticipation of being fed is the unconditioned response. There may be a historical parallel between the process and the effect. The also called unconditional stimulus is the active drug. Employing this terminology variant, the unconditional response is getting well. The conditional stimulus is the placebo. The conditional response would be getting rid of the disease because of learning by association. The problem with this argument is that it seems to imply that human beings have more control over their organism’s health than they factually do. On the bright side, there are other ways of interpreting classical and operant conditioning.

It is health that is real wealth and not pieces of gold and silver. Mahatma Gandhi, Indian Philosopher (1869-1948). Illustration: Megan Jorgensen.

References:

• Finniss, D. G., Kaptchuk, T. J., Miller, F. & Benedetti, F. (2010). Biological, clinical, and ethical advances of placebo effects. The Lancet, 375: 986-995.
• Mayberg, H. S., Silva, J. A., Brannan, S. K., Tekell, J. L., Mahurin, R. K., McGinnis, S. & Jerabek, P. A. (2002). The functional neuroanatomy of the placebo effect. The American Journal of Psychiatry, 159 (5): 728-737.Moerman, D. E. & Jonas, W. B. (2002). Deconstructing the placebo effect and finding the meaning response. Annals of Internal Medicine, 136 (6): 471-476