Technical Trading

Is the difference between upside and downside asset participation ratios relative to a benchmark a useful metric for evaluating asset investment performance? In his June 2014 paper entitled “On the Holy Grail of ‘Upside Participation and Downside Protection'”, Edward Qian defines and investigates the performance implications of the Participation Ratio Difference (PRD) as a measure of combined upside participation and downside protection. He defines the upside (downside) participation ratio of an asset as the ratio of expected excess return for the asset to the expected excess return of its benchmark when benchmark returns are positive (negative). “Excess” means in excess of the return on cash (such that cash has zero participation rates). He defines PRD as the simple difference between positive participation ratio (P+) and negative participation ratio (P-). He then investigates the relationship between asset PRDs and one-factor (market) alphas. He then checks PRDs for the S&P 500 sectors (with the S&P 500 Index as the benchmark) and PRDs for Russell style indexes (with the Russell 3000 Index as the benchmark). Using monthly returns of the S&P 500 index and its ten sectors during October 1989 through April 2014 and monthly returns of Russell broad and value-growth style indexes during January 1979 through April 2014, he finds that: Keep Reading

Does quantitative technical analysis work reliably in currency trading? If so, where does it work best? In their May 2013 paper entitled “Forty Years, Thirty Currencies and 21,000 Trading Rules: A Large-Scale, Data-Snooping Robust Analysis of Technical Trading in the Foreign Exchange Market”, Po-Hsuan Hsu and Mark Taylor test the effectiveness of a broad set of quantitative technical trading rules as applied to exchange rates of 30 currencies with the U.S. dollar over extended periods. They consider 21,195 distinct technical trading rules: 2,835 filter rules; 12,870 moving average rules; 1,890 support-resistance signals; 3,000 channel breakout rules; and, 600 oscillator rules. They employ a test methodology designed to account for data snooping in identifying reliably profitable trading rules. They also test whether technical trading effectiveness weakens over time. In testing robustness to trading frictions, they assume a fixed one-way trading cost of 0.025%. Using daily U.S. dollar exchange rates for nine developed market currencies and 21 emerging market currencies during January 1971 through July 2011, they find that:

Keep Reading

In which country stock markets is technical analysis likely to work best? In the October 2014 version of her paper entitled “Technical Analysis: A Cross-Country Analysis”, Jiali Fang investigates three potential cross-country determinants of technical trading profitability:

1. An individualism index, measuring the degree to which individuals integrate via cultural groups.
2. Market development and integrity metrics, including stock market size, stock market age, transaction costs and measures of investor protection, anti-director rights, ownership concentration and insider trading.
3. Information uncertainty metrics, including aggregate market turnover, volatility of cash flow growth rate and book-to-market ratio.

She considers 26 previously studied trading rules employing only past prices, classified as: variable moving average (VMA) rules, fixed-length moving average (FMA) rules and trading range break-out (TRB) rules. VMA rules are long (short) an index when a short-term moving average is above (below) a long-term moving average. FMA rules are similar to VMA rules, but hold a newly signaled position a fixed interval of 10 days. TRB rules generate buy (sell) signals when price rises above (falls below) the resistance (support) defined by prices over a specified past interval. Tests include both regressions and model strategies that are long (short) the market index as signaled and invest in the risk-free asset when there is no signal. Using cultural metrics, daily stock market index data and economic/financial variables for 50 countries during March 1994 through March 2014, she finds that: Keep Reading

A reader inquired about the validity of Martin Zweig’s Four Percent Model, which states (from pages 93-94 of the 1994 version of Martin Zweig’s Winning on Wall Street):

“The Four Percent Model for the stock market works as follows. First, It uses the Value Line Composite Index…an unweighted price index of approximately seventeen hundred stocks… All you need to construct this model is the weekly close of the Value Line Composite. You can ignore the daily numbers if you wish… This trend-following model gives a buy signal when the weekly Value Line Index rallies 4% or more from any weekly close. It then gives a sell signal when the weekly close of the Value Line Composite drops by 4% or more from any weekly peak. …That’s all there is to it. …The model is designed to force you to stay with the market trend.”

We execute this description as follows (after identifying the first signal):

• After a buy signal, generate the next sell signal upon a 4% or greater decline from a subsequent high water mark (including the buy signal level).
• After a sell signal, generate the next buy signal upon a 4% or greater advance from a subsequent low water mark (including the sell signal level).

We test the usefulness of the signals on the following exchange-traded funds (ETF) over their entire available histories: SPDR S&P 500 (SPY), PowerShares QQQ (QQQ), iShares Russell 2000 Index (IWM) and Guggenheim S&P 500 Equal Weight (RSP). Using weekly closes of the Value Line Geometric Index and the dividend-adjusted weekly opens of the selected ETFs from their respective inceptions through September 2014, we find that:

Keep Reading

Do stock pairs that track in the past reliably track in the future? In his January 2014 paper entitled “On the Persistence of Cointegration in Pairs Trading”, Matthew Clegg assesses the persistence of cointegration among pairs of liquid U.S. stocks. Specifically, he investigates whether pairs of equities that are cointegrated in an initial interval are likely to be cointegrated in a subsequent interval. He uses calendar years as initial intervals and focuses on next years as subsequent intervals. He also considers shorter subsequent intervals. He employs a variety of methods to measure pair cointegration to ensure robustness of findings. Using daily returns for constituents of the S&P 500 (as of August 13, 2013) during January 2002 through December 2012, allowing ten years of persistence tests, he finds that: Keep Reading

Do financial markets adapt to widespread use of an indicator, such as Bollinger Bands, thereby extinguishing its informativeness? In the August 2014 version of their paper entitled “Popularity versus Profitability: Evidence from Bollinger Bands”, Jiali Fang, Ben Jacobsen and Yafeng Qin investigate the effectiveness of Bollinger Bands as a stock market trading signal before and after its introduction in 1983. They focus on bands defined by 20 trading days of prices to create the middle band and two standard deviations of these prices to form upper and lower bands. They consider two trading strategies based on Bollinger Bands:

1. Basic volatility breakout, which generates  buy (sell) signals when price closes outside the upper (lower) band.
2. Squeeze refinement of volatility breakout, which generates buy (sell) signals when band width drops to a six-month minimum and price closes outside the upper (lower) band.

They assess the popularity (and presumed level of use) of Bollinger Bands over time based on a search of articles from U.S. media in the Factiva database. They evaluate the predictive power of Bollinger Bands across their full sample sample and three subsamples: before 1983, 1983 through 2001, and after 2001. Using daily levels of 14 major international stock market indexes (both the Dow Jones Industrial Average and the S&P 500 Index for the U.S.) from initial availabilities (ranging from 1885 to 1971) through March 2014, they find that: Keep Reading

Can investors exploit monthly persistence in the value premium for U.S. stocks? In his February 2014 paper entitled “Exploiting Factor Autocorrelation to Improve Risk Adjusted Returns”, Kevin Oversby investigates whether investors can exploit the fact that the Fama-French model high-minus-low (HML) value factor exhibits positive monthly autocorrelation (persistence). The HML factor derives from the difference in performance between portfolios of stocks with high and low book-to-market ratios. Prior published research indicates that the value premium concentrates in small firms, so he focuses on stocks with market capitalizations below the NYSE median. His test strategies each month invest in capitalization-weighted small value (small growth or small momentum) Fama-French portfolios when the prior-month sign of the HML factor is positive (negative). The strategies additionally retreat to a risk-free asset (such as U.S. Treasury bills) if the prior-month return for the test strategy is negative. Using HML factor values and monthly portfolio returns for small value, small growth and small momentum Fama-French portfolios, he finds that: Keep Reading

Do hedge fund managers who use technical analysis beat those who do not? In their May 2014 paper entitled “Sentiment and the Effectiveness of Technical Analysis: Evidence from the Hedge Fund Industry”, David Smith, Na Wang, Ying Wang and Edward Zychowicz examine the relative performance of users and non-users of technical analysis among hedge fund managers in different sentiment environments. They hypothesize that short-selling constraints prevent market correction of mispricings when sentiment is high (overly optimistic), but not when sentiment is low (overly pessimistic). Discovery of mispricings via technical analysis may therefore be more effective when sentiment is high. To test their hypothesis, they compare the performance of hedge funds that report using technical analysis to that of hedge funds that do not, with focus on the state of market sentiment. They define the market sentiment state as high or low depending on whether the monthly Baker-Wurgler market sentiment measure is above or below its full-sample median. Using end-of-period status on use/non-use of technical analysis and monthly returns for 3,290 live and 1,845 dead funds from the Lipper TASS hedge fund database and monthly market sentiment data during January 1994 through December 2010, they find that: Keep Reading

Does technical market analysis work? In their June 2014 paper entitled “Technical Market Indicators: An Overview”, Jiali Fang, Yafeng Qin and Ben Jacobsen examine the profitability of 93 market indicators as applied to the S&P 500. Of the 93, 50 are market sentiment indicators that attempt to predict market behavior based on the supposition that stock prices tend to rise (fall) when bullish (bearish) sentiment dominates. The remaining 43 are market strength indicators that attempt to predict market trend continuation based on breadth of movements as indicated by volume, number of advancing/declining issues and number of periodic highs/lows. 65 of the 93 indicators are raw (such as numbers of advancing and declining stocks per day), and 28 involve measures constructed to suppress noise (such as number of advancing issues minus number of declining stocks). The authors use the S&P 500 as a test market because of its long history. They consider entire sample periods, equal subperiods, different economic regimes (expansion or contraction) and different sentiment regimes (bullish or bearish as indication of degree of investor irrationality). They employ a generous 10% significance level for statistical tests, with and without estimated trading frictions of 0.10% for switching between the market and a risk-free asset. Using the longest samples available for each indicator through the end of 2010 or 2011 (averaging 54 years and as long as 200 years), they find that: Keep Reading

In reaction to “10-month Versus 40-week Versus 200-day SMA”, a reader inquired whether using measurement intervals of longer than a month to calculate simple moving averages (SMA) would suppress trading compared to monthly intervals and thereby lower trading frictions and improve performance. To check, we compare the performance of simple moving averages based on 12 months (SMA12M), six bi-months (SMA6B) and four quarters (SMA4Q). SMA6B samples six data points bimonthly, with each measurement spanning 11 months. SMA4Q samples four data points quarterly, with each measurement spanning 10 months. Using monthly dividend-adjusted closes for SPDR S&P 500 (SPY) from inception in January 1993 through April 2014 (about 21 years), along with the contemporaneous monthly 3-month Treasury bill (T-bill) yield, we find that: Keep Reading