To be a successful trader one should be part data scientist. Although there are some highly successful traders that hinge their plan on subjective analysis artforms, the road to long-term profits in active trading should be grounded in the laboratory of data. The simple reason for this: failure, as much as success, must be measured and understood because market outcomes are often inherently volatile and unpredictable. The scientific method allows us to test trading hypothesis, learn from mistakes, and quantify risk. Systematic traders understand that without the integrity of data science, they are simply ticker-tape cowboys.
Having an analysis framework is an important departure point for the systematic investor. That framework could be fundamental / value analysis - translating measures of intrinsic value into trading signals. An example of this would be the Dogs of the Dow trading strategy. For market technicians core intrinsic value relationships are too complex to model completely. Instead, a technical analyst focuses on patterns of supply and demand for an investment instrument. Those patterns of subjective market valuations are revealed in the price and volume of every stock.
“It’s not good enough to be anecdotal or doctrinaire when it comes to trading.”
Market technicians believe in the market’s message. They construct price and volume charts to read the tea leaves, so to speak, about the future direction of a market price. But here’s where this backward-looking artform often fails: how can a technical analyst place any faith in the reading of these charts? It’s not good enough to be anecdotal or doctrinaire when it comes to trading. It’s not good enough to show a tidy chart that reveals, for example, a head and shoulders pattern, and assert a projection for future price movement if there is no data from which to develop a measure of confidence in the prognostication.
There should be some standard of evidence to support a particular chart reading. If there is no evidence, there’s no foundation for acceptance. A technical trader should quantify the probabilities of future outcomes. Data science allows the diligent systematic investor to develop a level of confidence in a market environment that is fundamentally uncertain. Risk must be quantified!
The following statistical models offer two different approaches to translating the Stock Trends data into sample spaces and predictions of future returns or price performance.
Stock Trends Inference Model
How does Stock Trends help us turn stock market data into actionable quantitative measures of confidence? First, the Stock Trends indicators are by definition categorical - they translate market price and volume data into factor variables or independent variables. In a data science setting, we can use these independent variables as inputs and measure a relevant outcome, or output. The significant outcome for investors, of course, is the future return. If a dependent relationship is established between the inputs and the output, the trader can measure a confidence level for the desired trading outcome. Let’s now look at each of the Stock Trends indicators and how they fit our Stock Trends Inference Model.
The Stock Trends trend categories are the result of a method of translating market price data (quantitative variables) into categorical variables. For instance, the June 19, 2015 closing price of Apple (AAPL) was $126.60. The Stock Trends trend indicator categorizes that price by applying a framework for qualification and putting the current price into a long-term price context. Using 13-week and 40-week average prices as guideposts, the trend indicator - now Stock Trends Bullish () - gives us a factor variable for the $126.60 market price.
A base test, then, would be to measure how a market price performs when it is in this trend category. However, we would want a more granular categorization because within each trend category there are many ancillary variable qualifications. For instance, a Bullish trend category can be relatively new, or it can be quite entrenched.
That is why Stock Trends publishes trend counters. They give us a better understanding of the time frame of the trend category. In our Apple example, we can see that the Bullish trend category was in place for 92-weeks, about twice the average length of a typical Bullish trend and that the strong Bullish indicator had been in place for 22-weeks. So now we can ask the following question: how have stocks performed when they have been in a Bullish trend category for about 92-weeks, and also in a strong Bullish indicator for the most recent 22-weeks?
But our granularity can be improved even more. We also recognize that within any trend there are varying levels of price momentum. Stocks rally and retreat. The Stock Trends Relative Strength Indicators provide us with a method for translating price performance into factor inputs. The 13-week RSI values are discrete variables that can be cut into bins of specific ranges of values. By qualifying each stock’s trend by its relative price momentum to the broad market we can now be more specific about the characteristics we are sampling. In the case of our Apple example, its 13-week RSI was 100. This indicates the stock is only performing at par with the S&P 500 over the previous 13-weeks. Now we can sample for Bullish stocks that also share this condition.
The RSI +/- indicator is a binary signal of whether a stock has outperformed or underperformed the broad market in the past week. Again, this indicator can be used as another factor input. Apple underperformed the S&P 500 index the previous week and therefore has a (-) indicator.
Finally, another factor variable that Stock Trends creates is derived from the weekly volume of shares traded. Three different factor levels characterize the weekly volume so that we can differentiate stocks further by which level the trading volume fits. The example shows that Apple had neither high nor low volume of trading, so its volume can be characterized as normal.
With these composite factor variables, published in each Stock Trends Report, the Stock Trends Profile presents the results of the Stock Trends Inference Model. In the case of Apple, shown below, we can see that the current Stock Trends indicators are relatively positive: the future 4-week return of Apple has a 57% probability being higher than the expected mean random return of a stock, which is 0%. Remember, that a randomly chosen stock has a 50% chance of having a 4-week return greater than 0% (see The random outcome benchmark). The model shows that AAPL has a 62.2% chance of besting the base mean 13-week random return, which is 2.19%, and a 56% probability of besting the mean 40-week random return (6.45%).
These probabilities might not strike you as significantly positive. However, they do indicate that the trend and momentum conditions for AAPL are sufficiently supportive of a continued bullish stance for Apple investors. The analysis also tells us that AAPL is more appealing than stocks with lower return expectations. You can compare the returns expectations of industry stocks in the associated heatmap that ranks the expected future returns.
As it turned out for this particular example AAPL did outperform the base random return in the next 4-week period with a return of 2.4%, however, the stock retreated subsequently and was -10.4% after 13-weeks and -16.5% after 40-weeks (post-June 16, 2015 observation). This affirms that the model's probability statements do not present any certainty in the predictions, and should not be construed otherwise.
This is the analysis framework of Stock Trends: translating the weekly trading statistics of an issue into factor input variables. It is how we interpret these variables and their significance in predicting future price performance that makes Stock Trends a unique and effective data science application. The Stock Trends Inference Model statistically measures the change in stock price that follows from each market condition defined by the composite of the inputs of each Stock Trends indicator combination.
Stock Trends covers the North American stock market - thousands of issues every week are categorized by the Stock Trends indicators. Each of these observations since 1980 - now numbering over 11-million records - can be used as input variables in models that measure the subsequent price change in the categorized stock. We can ask the question: what kind of returns did a stock have after it was categorized by the Stock Trends indicators? Do stocks that have a Bullish trend indicator and high price momentum perform better, on average, than other stocks? Is there any statistical evidence that momentum trading is profitable? Does a Bullish Crossover offer a good trade entry signal? More broadly, does the data support many of the doctrinaire positions of technical analysis? The Stock Trends Inference Model attempts to answer these types of questions.
"Every technical analyst who presents a price chart as evidence of a buy signal must also present a distribution graph of the expected returns. If they don't, take their advice with a grain of salt."
Stock Trends analysis framework is simple but specific. It looks at certain important aspects of technical analysis - trend and price momentum. Another analysis framework might be centered on other algorithms of price and volume, and on a different time frame. Every investor has to choose what analysis framework fits their own assumptions about the dependent relationships in the market. However, each analysis framework must be measurable. The litmus test of this measurement should be the presentation of data, the display of returns distributions. Indeed, in my opinion, every technical analyst who presents a price chart as evidence of a buy signal must also present a distribution graph of the expected returns. If they don’t, take their advice with a grain of salt. Your success as a systematic investor will reflect your diligence in making data science integral to your trading strategies.
Stock Trends RSI +/- Pattern Analysis Model
The Stock Trends RSI +/- Pattern Analysis Model generates weekly probability analysis of the named sample space for thousands of stocks and ETFs. The model measures weekly alpha (market outperformance) for North American equities and generates statistical inference of expected returns from the pattern portfolios. The end result is a Profit Factor for a given pattern of outperformance/underperformance which short-term traders can use in their trading strategies and trade setups.
The Stock Trends RSI +/- indicator is a simple binary indicator of market outperformance or market underperformance on a given week. If a stock issue or ETF has outperformed the market benchmark - the S&P 500 Index for the U.S. market and the S&P/TSX Composite Index for the TSX - a plus sign (+) is assigned. If the stock or ETF underperforms the benchmark a minus sign (-) is assigned. This indicator is described in the Relative Strength Indicator section of Chapter 4 - Guide to Stock Trends Symbols and Indicators.
Here is a sample history of the indicator for Ford Motor Company (F):
The RSI +/- Pattern Analysis Model asks the question: in a given trend category, does the pattern of the weekly market over/underperformance indicate future market over/underperformance, and what is the expected return of the overperformance/underperformance? If a stock outperformed the market last week, will it outperform the market this week? If it outperformed the market for four consecutive weeks, what is the probability it will outperform the market in the following week? If so, what is the expected return? These are questions traders ask that all relate to a core gambling challenge about winning streaks. The binary nature of the Stock Trends RSI +/- indicator provides us with a sample space similar to a coin toss.
Generally, if we examine the probability of market outperformance/underperformance on a weekly basis - and this will be true for any period given enough data - there is a near 50 percent probability of either outcome. As an example, Caterpillar Inc. (CAT) has outperformed the S&P 500 index on 50.2 percent of trading weeks since March 21, 1980. Bank of America (BAC) has done so 48.8 percent of trading weeks, Walmart (WMT) 51.6%, Microsoft (MSFT) 52.5%, Intel (INTC) 51.9%, and Coca-Cola (K) 50.6%. Across all North American stocks, 48 percent of stocks, ETFs, and indices have outperformed their relevant benchmark on a weekly basis when we look at the Stock Trends 11.25-million records of weekly market outperformance/underperformance. Although it’s not a perfect or fair coin, the binary outcome of the RSI +/- indicator tends toward the distribution of randomness.
But does the probability of outperformance vary with a given price trend? Does previous outperformance indicate future outperformance? If we have tossed three heads in a row, will the next toss yield a head? In a true random toss, of course, the probabilities remain the same - 50% - regardless of the streak. But does the market present us with an opportunity to measure outperformance and identify imbalances where the market trend tilts probable outcomes away from long-term randomness toward opportunistic trading with an unfair coin? Are there times when a pattern of outperformance/underperformance turns away from the house and more favorably to the investor?
I first introduced the RSI +/- Pattern Analysis Model in a previous editorial, but let’s look at another example of the model at work and show how a trader can possibly use it. In the Ford (F) example shown above, we can see that the most recent pattern for the RSI +/- indicators which tells when F outperformed or underperformed the S&P 500 index each week. Since 1980, F has outperformed the market in 47.3 percent of the weekly trading. However, the RSI +/- Pattern Analysis Model puts that relative market performance metric in the context of price trend.
In this example, on April 26, 2019 Ford's stock (F) is in a Stock Trends Weak Bearish () trend. Its long-term trend category is still Bearish, but the price momentum since the stock’s low at the end of 2018 is indicating a trend reversal. F has been Weak Bearish for 4 weeks now and the stock is outperforming the S&P 500 index by 6 percent (13-week Stock Trends RSI 106) thanks to the post-earnings move last week, so we know that there is a growing buy conviction for this automotive play.
Focusing just on the current trend category, Weak Bearish, and the recent pattern of weekly outperformance/underperformance, do previous observations of similar patterns in previous Weak Bearish categories tell us that the probability of future market outperformance has been enhanced? And can it tells us what the expected return will be? The defined sample space and statistical inference provide us with the answer.
The stock has outperformed the S&P 500 index for six consecutive weeks. That’s a pretty good winning streak. The model looks to historical weekly data since 1980 looking for similar streaks where F was in a Weak Bearish trend and also had such a consecutive string of weekly market outperformance, measuring post-observation returns. Here is the result of the analysis, which subscribers to Stock Trends Weekly Reporter will find at the bottom of the Profile section of the Stock Trends Report:
The analysis is an inference model, so a minimum number of observations is required for the sample - 20. In this case, the required number of observations is met up to the 6-week pattern, which is consecutive market outperformance (++++++).
The RSI +/- Pattern Analysis shows that previous observations of one week patterns of market outperformance (+) with the stock in a Weak Bearish trend indicator numbered 189, with 128 of those observations (67.7%) followed by another week of market outperformance. Further, when this event happens the estimated mean return (average) for the population (statistical inference of the mean derived from the sample) is within a range between 4.3% and 5.9%. Of course, the flip side of the coin shows that 32.3% of the observations were followed by market underperformance the following week and that the mean return of this population is between -2.2% and -0.9%. The statistical inference of the sample and its probability space tell us that the current trend market outperformance of F indicates a positive bias toward positive returns in the following week.
Remember, the occurrence of market outperformance overall trends for F is measured at 47.7%, below the random observation marker of 50%. The 20 percent difference in probability of seeing market outperformance in F is significantly higher than the observed frequency seen broadly in over 2,000 weeks since 1980.
If we look at patterns of F’s relative performance to the market going back several weeks, the analysis shows the sample statistical inference result for the previous six weeks. It just so happens that the stock has just outperformed the market six weeks in a row. The model seeks to answer the question: what are the probabilities that it will outperform the market in week seven and what would be the expected return?
The sample of this observation, a streak of six weeks of market outperformance (++++++) where F is in a Weak Bearish trend is obviously smaller, but we can use statistical inference to estimate the population returns for both sides of the event - market overperformance and market underperformance the following week.
The analysis shows that there is a 68% probability that F will outperform the market again and that the expected return in the upcoming week is between 3.7% and 7.1%. Alternatively, there is a 32% probability of F underperforming the market with an expected return between -7.8% and 0.3%.
How do we compare and evaluate this analysis? Like any portfolio of returns we must measure the gains relative to the pains, so to speak. Investors should be aware of the Profit Factor as one measure of this relationship. Stock Trends presents the Profit Factor for all Stock Trends trading strategies, and you can find these values on the Stock Trends strategy description pages (example Stock Trends NYSE Portfolio #1). Basically, the Profit Factor calculates the ratio of the probability of profits from a winning trade and the probability of losses from a losing trade. This ratio tells us if the rewards of being right outweigh the penalties of being wrong. The higher the Profit Factor, the more the rewards of winning trades outweigh the cost of losing trades.
Here the RSI +/- Pattern Analysis Model provides a Pattern Profit Factor. Given the observation described for F (++++++, in a Weak Bearish trend currently), the Pattern Profit Factor is 3.1. You can see how that relatively good measure stacks up against the RSI +/- Pattern of other stocks and ETFs by referring to their Stock Trends Report pages under the Profile tab (always under the Stock Trends Inference Model section).
Traders could use the RSI +/- Pattern Analysis in conjunction with the Stock Trends Inference Model, which presents longer-term estimated future returns (4-week, 13-week, 40-week), to either confirm entry timing or develop other more complex options setups for short-term trades. The RSI +/- Pattern Analysis addresses weekly expected returns and probable alpha outcomes. Going long stocks and ETFs with high Pattern Profit Factors and shorting those with a low Pattern Profit Factor is a naive approach, but more refined setups are advised.