Raymond Rondeau leads a class in AAII's new Essential Investing Video Course. Go to https://www.aaii.com/ves for more information and to subscribe.
We are starting a new series of articles to the CI lineup, covering the Classic Technical Indicators. Writing this series is Raymond Rondeau, a fairly well-known name in the field of technical analysis and one that is familiar to our AAII members. Ray has been involved with AAII for the last 10-plus years in various capacities, including writing for the AAII Journal, serving as the AAII Boston Chapter president and speaking at more than 20 local AAII chapters. Ray also lectured at the AAII Investor Conference last year (2015) in Las Vegas. More information about Ray can be found here.
With Ray’s first article scheduled for release in June of 2016, I thought that it would be a good idea to interview him on this upcoming series.
Because of the length of the interview, we are separating it into two parts. Here in Part 1, we discuss the many potential negatives and pitfalls of working with indicators and technical analysis in general. But be sure to look for Part 2 of the interview in the May CI, where Ray covers the benefits of working with technical indicators. This two-part interview serves as a good introduction to Ray’s new CI series called Classic Technical Indicators, which will debut in the June issue of CI.
Jackie: Let’s start off with what motivated you to write this series on the “classic technical indicators”?
Raymond: Certainly part of the reason is a chance to work with you and continue my relationship with AAII. The other reason I decided to do this series is that in my 15-plus years of working with individual traders and investors, I find that there is a lot of misconceptions and misuse in regard to the interpretation and implementation of these “classic indicators.” In fact, I have seen some cases where an investor probably would have been better off if they had not been attempting to utilize indicators at all, simply because they attempted to use them inappropriately and in a way that they were not designed for.
Jackie: Could it be that the indicators themselves are at fault?
Raymond: No, the indicators themselves are not at fault, but they are being used by people, who are and act human. We, as humans, have a number of evolutionary (nature) traits that can often hurt us when investing. The two most commonly known and referenced traits that can lead us astray are fear and greed. Additionally, we as humans also develop (nurture) a host of other biases and emotional components along life’s path, which can affect how we see all things. Or, as Anais Nin stated, “We don’t see things as they are, we see them as we are.”
In relation to indicators, these factors can affect how indicator signals are interpreted and the amount of relevancy that we attribute to them. The good news is that indicators can help us overcome some of the instinctive and inherent human limitations that we all have when investing in the markets. The bad news is that some newer investors can fall into the trap of becoming overdependent and overconfident when interpreting indicator signals.
Jackie: Can you give us some specific reasons why an investor could be worse off by using indicators?
Raymond: I have found that some beginning investors can become so overdependent on a technical signal that it can become counterproductive. If an investor becomes too dependent, they can stop searching for or considering other more important pieces of relevant information that should be factored into the overall analysis. A single indicator signal does not mean one should not evaluate other criteria such as fundamentals, industry performance, insider activity, derivative readings and other more relevant technical signals from the charts. When an investor starts using indicators, it’s important to make sure that they don’t start to get tunnel vision, which is rarely a good thing in the markets.
Another potential problem that can arise with those who start to use technical analysis and indicators is over-focus. This can best be explained by the statement “out of sight out of mind.” Sometimes when an investor has something tangible to focus on, they actually become more intellectually and emotionally involved. Because of this, they become more inclined to take action, which can lead to over-trading and a portfolio that becomes out of line from their originally targeted allocations.
Jackie: Are there any other concerns beyond overdependence and overfocusing?
Raymond: Equally concerning is when a new investor starts to become so overconfident in an indicator that they start increasing their risk. In other words, they start taking on larger positions than they normally would have, had they not received the concurring indicator signal.
I don’t care how many indicators are giving you a bullish signal or how great of a technical setup your charts are showing. The price of the “issue” is going to do what it is going to do, and this is based on numerous other non-technical factors. Fundamentals, new innovations, Federal Drug Administration (FDA) and litigation decisions, economic, geographical or political factors and many other macro-like influences are the driving forces behind long-term price movements, not an indicator.
Indicators should not be interpreted as Nostradamus-like prophecies. At one time, Enron and Bear Sterns were displaying strong technical oversold readings, but they still are not worth anything today. The acknowledgment that other factors are the dominant driving forces behind price movements is an important concept for a technician to keep in mind. With this acknowledgment one is less likely to become overly confident in any technical based or single indicator reading.
Jackie: Earlier, you referred to implementation as a potential problem. What did you mean by that?
Raymond: In my opinion, an indicator is just a secondary tool to direct price analysis and consideration of the many other company and macro factors that I just mentioned. I think that most technicians would agree that indicators are best used as a complementary piece of the entire appraisal process. Indicators provide us with an alternate or additional way of evaluating a piece (or pieces) of information. Indicators were not designed to be used in isolation. They are best used to assist in clarifying and giving us additional insights to the primary price action.
Unfortunately, I often see many investors looking to (or for) a single indicator to be their Holy Grail, or sole determinant. They seem to be looking for that one-step solution for determining probable future price direction. Of course it would be nice if there were one specific indicator and set of static parameters that worked across all asset classes and in all cases. If this were true, we could all save ourselves a lot of trouble, but unfortunately logic reveals that this is not feasible and experience reinforces that this is not attainable.
Jackie: How then does an investor make an accurate appraisal about what an indicator is really showing?
Raymond: I would say by understanding not only the measurements the indicator is referencing (i.e., closing price, volume) but also the method of the calculations. In truth, most indicators are just static algorithms that are being calculated from just one or two pieces of information. And when one understands the mathematics behind the indicator, it tends to remove the “rose-colored” magical prophetic associations with it. This of course can help counter the overconfidence/overdependence/overfocus problems that we alluded to earlier.
This is why I plan on spending some time in each article covering not only how the values are calculated but how the indicator is actually constructed. Understanding the mathematics can give the user deeper insights into some of the pros and cons of that specific indicator and when it might best be applied or filtered in various circumstances.
For instance, how would a big dividend payout and the subsequent drop in the securities trading price affect the readings of a shorter-term oscillator from a designated overbought area? How would a stock split or reverse split on a price-adjusted chart affect various quantitative indicator levels and their interpretations? How would arithmetic versus semi-logarithm chart settings affect the visual perceptual interpretation of a changing price/indicator divergence slope?
Jackie: In addition to implementation concerns, you also mentioned problems with indicator interpretations. What are examples of those?
Raymond: Yes, it is kind of on the same theme. I have found that some novice investors often try to apply the wrong indicator(s) to not-the-most-advantageous situations. For instance there are various types of indicators. Some indicators, and the ones that I will be focusing on in this Classic Technical Indicator series, are more appropriate for individual security analysis. Oddly, there are some indicators that won’t work on specific individual securities. Indicators like put/call ratios and implied volatility are two obvious examples, due to the fact that there might not be any listed option(s) and/or very minimal activity for a certain security.
Then there are indicators that don’t work on any individual securities and that focus solely on overall market health such as: advance/decline line, VIX (CBOE Volatility Index) and market breadth. Interestingly with some of these market-based indicators, their entire approach and interpretation of the data can be completely different. Many popular indicators can work effectively on both the market indexes and individual securities, such as moving averages, moving average/convergence divergence (MACD) and relative strength index (RSI).
Jackie: With all of the different indicators and types, are there ways to combine them to be more effective?
Absolutely, in fact an important relationship in technical analysis is the belief that indicators often work better when combined. When indicators are combined with (and in the right combination) other indicators, they tend to work synergistically.
This reinforces another general principle of technical analysis: The more confirming technical signals one has, the better. The greater the number of concurring and reinforcing signals being displayed, the more weight and influence one may want to attribute to the combined readings.
Jackie: New investors are often confused when attempting to use technical indicators for the first time and do not know where to start. Is this because there is actually too much coverage and information on too many indicators?
Raymond: Certainly, with today’s technology there is a lot of information available on all the indicators. This is not necessarily a bad thing. To me, the problem lies in that there are too many people trying to “create their claim to fame” by altering the methodology and/or interpretations of each of the indicators. This is a concept that I routinely cover in my presentations in regard to technical price (and candle) patterns.
Again, in my opinion, you don’t need to all know all 150 price patterns to be successful or to gain an edge as an investor. Creating a large number of entities may be helpful to sell books and investment courses, but if you ask most technicians, they will probably tell you that you just need to be able to recognize the most relevant and effective ones. Focusing on the half-dozen or so primary reversal and continuation price patterns (and key indicators) should be more than enough for most investors. The efficient investor wants to selectively watch for the key signals that are going to get the market’s attention and that are likely to cause money to flow off of them. When working with charts and indicators, we want to be sure and try and avoid the “analysis paralysis” trap as this is a very real and legitimate concern.
In fact, one of my goals with this series is to present the information on indicators in a succinct and efficient manner for easy assimilation and application. I want to provide each reader with enough statistics and information so that they can make an informed choice on which few indicators that they feel are best for them. Strategically selecting only a few indicators that are tailored to each individual investor’s circumstances can help diminish this aforementioned “information overload” concern.
Jackie: Continuing on this theme: Is this the main reason that some newer investors lose interest with using indicators and eventually give up?
Raymond: I am sure that there are many reasons. It’s important to remember that when we are utilizing technicals, we are dealing with improving probabilities, not attaining guarantees. We all desire and expect to be rewarded for our efforts and when we aren’t, it is only natural to want to look elsewhere. I am sure this is a factor to some extent but, unfortunately, it is a reality that all investors must accept.
Now there is one reason that many abandon technicals that we don’t have to accept, because it is a misconception. And that is the perceived direct correlation between indicator effectiveness and complexity, which in my opinion is not necessarily true. I feel that this is a myth, and one that is often being propagated and propelled by others in the investment arena for potential monetary gain. I can certainly understand the allure of trying to attain higher profile, more complex methodologies and esoteric indicators, but unfortunately the return expectations seldom equate to the reality.
Jackie: So are you advocating the KISS method (“Keep It Simple, Stupid”) with using indicators?
Raymond: I believe Einstein once stated: “Make things as simple as possible but not simpler.” I am not advocating a simple or complex approach, but the most effective approach. An intellectual, diversified and disciplined-based approach to investing makes the most sense to me.
What I am “advocating”—or should I say, what I have learned through my years of evolution as a trader—is that more complex is not necessarily more effective, often it is just more complex. This follows the Occam’s razor idea that the simpler of competing explanations is to be preferred. I run an investment group for advanced traders that meets regularly at MIT in Boston and they always want to make things more complex but more often than not they find after exhaustive efforts that the simplest approach is the best.
Jackie: How does an investor gauge the line between indicator complexity and effectiveness?
Raymond: The computer scientist Grace Hopper once stated that: “One accurate measurement is worth 1,000 expert opinions.” Certainly, one seemingly logical approach would be to do a properly performed and thorough backtest/forward-test. Now let me state right off that this method, like almost everything in the investment world, is far from perfect. It has many very real tangible limitations, and “past performance is absolutely no guarantee for future results.”
Markets change and markets evolve, as do individual securities’ price movement characteristics, but at some point we have to make assumptions and start somewhere to formulate an opinion. Let me ask you this: Say you had Indicator “A” that had a 20+ year long-term consistent track record of outperforming a buy and hold approach on a stock or index. Then say you had Indicator “B” that had a 20+ year long-term consistent track record of underperforming a buy and hold approach on a stock or index. Which signal would you be most likely to follow?
Jackie: Isn’t attaining this quantitative information difficult and/or expensive?
Raymond: That brings us back to my series. Yes, to make this assessment we do need to do the computations. This is going to be a primary feature of my Classic Technical Indicator series and it will be one of the key elements that separate it from the many other purely theoretical-based articles published. I am going to use one of the most advanced indicator/strategy evaluation programs available in TradeStation’s Portfolio Maestro to compute the statistics on hundreds of symbols, over various strategically chosen long-term periods, to see if an indicator really would have been effective, at least in the past. Then I will further exploit the advanced capabilities of the Portfolio Maestro program by creating optimized data tables to evaluate the results of various possible settings and parameters.
By displaying the information in this manner, it will give each reader the opportunity to decide for themselves whether they want to use the indicators traditional settings or adjust them to the more “historically” verified effective settings. Furthermore, I am also planning on providing a variety of different readings. Presented this way, readers will be able to focus on the columns of information that they find most important to evaluate and that matches their own personal investment style. This will include elements such as total profitability, win/loss ratios, Sharpe ratios, number of trading investment signals per time period, maximum drawdowns etc.
Portfolio Maestro is a highly sophisticated tool from TradeStation that provides portfolio-level performance reporting, risk assessment and optimization for virtually any combination of symbols and strategies. More information can be found at the TradeStation website.
Jackie: Earlier you referenced the changing characteristics of an individual security. Does this mean that you advocate adjusting or contouring indicator parameters over time?
Raymond: This is a grey area and, to me, one that I feel that should be left up to each individual to decide. I don’t think that many experienced investors would question the concept that the price movement characteristics of a company can change. For example: I don’t think it is a stretch to assume that the magnitude and volatility of Microsoft stock prices are different now (as a blue-chip behemoth) than when it was an initial public offering (IPO) in 1986. Naturally, if one agrees with this premise, then an indicator selection and/or sensitivity adjustments would seem logical as a company’s lifecycle progresses over time.
I do know many investors who adjust or self-optimize settings on issues, and they report that it is very effective. Furthermore, there are programs available like OmniTrader that not only optimize settings, but also backtest a large number of indicators (called “strategies”) on a specific issue, to see which actual indicators (strategies) had performed the best.
Jackie: You mentioned limitations in regard to the backtesting. Can you expound on this a little for those readers that have no experience with backtesting?
Raymond: First, I think that it is critical for investors to realize that, much like most things related to the markets, even if you do everything perfectly, there is absolutely no guarantee of success. This is something that is certainly true with indicators in general and with backtesting specifically. All we can do as investors is to focus on what we can control. In regard to backtesting, all we can do is to try to avoid many of the common problems associated with it. For these articles, I will be overseeing and adjusting the computation process to avoid as many of these common pitfalls as possible. Some common pitfalls include: backtesting on too few symbols, backtesting over a time period that is too short, backtesting over a biased time period, misinterpretation of back testing results and misinterpretation of optimized parameter results. The box below explains these in more detail.
Common Pitfalls When Backtesting
Backtesting on too few symbols
To judge if an indicator in itself is truly effective, we would like to measure its performance over a multitude of symbols covering various sectors and industries.
Backtesting over a time period that is too short
Here we want to see a good consistent long-term track record. As in basic probability theory (law of large numbers), results from a greater number of trials should approach more accurate or expected results.
Backtesting over a biased time period
Here we want to eliminate a market directional bias by back testing over (long) time periods, where the market has shown a zero or neutral return. Note: We will be calculating and posting results on indicator performance in both bullish and bearish market environments for comparative analysis.
Misinterpretation of backtesting results
Here we want to be sure that we are properly analyzing our results. We want to see results displaying predictability and consistency, as opposed to peak to peak anomalies or other highly skewed readings. For instance, we want to see a good sustainable and consistent long-term upward sloping linear profit / loss curve. We don’t want to see, myopic and skewed quantitative bottom line results on an indicator that had one or two great but brief periods.
Misinterpretation of optimized parameter results
As Twain once stated: “There are… lies, damned lies, and [then there are] statistics.” We definitely don’t want to start curve fitting or distorting results to attain an illusionary satisfactory result. If we’re going to try and optimize parameters (say a 175-period moving average versus a 200-period moving average), then we want to be sure that the optimization tables are showing a Gaussian-like distribution curve. We don’t want to be choosing purely random, isolated mathematical outperforming anomaly periods such as a 178-day moving average. In this case, we would want to see supporting data on near proximity measurements (170 – 177) & (179 – 186) to support and substantiate the selection.
Jackie: This is the end of Part 1 of our interview with Raymond Rondeau. Part 2 of this interview, where Ray delves into the benefits of using technical indicators, will appear in the May issue of CI.
Discussion
FREE REPORT
Charles Livingston from TX posted over 10 years ago:
Leonard Eveleth from FL posted over 8 years ago:
You need to log in as a registered AAII user before commenting.
Log InCreate an account