Technical Trading

How does stock pairs trading performance interact with lagged pair correlation and volatility? In her May 2016 paper entitled “Demystifying Pairs Trading: The Role of Volatility and Correlation”, Stephanie Riedinger investigates how stock pair correlation and summed volatilities influence pair selection, pair return and portfolio return. Her baseline is a conventional pairs trading method that each month: (1) computes sums of daily squared normalized price differences (SSD) for all possible stock pairs over the last 12 months and selects the 20 pairs with the smallest SSDs; (2) over the next six months, buys (sells) the undervalued (overvalued) member of each of these pairs whenever renormalized prices diverge by more than two selection phase standard deviations; and, (3) closes positions when prices completely converge, prices diverge beyond four standard deviations, the trading phase ends or a traded stock is delisted. A pair may open and close several times during the trading period. At any time, six pairs portfolios trade simultaneously. She modifies this strategy to investigate correlation and volatility effects by: (1) measuring also during the selection phase return correlations and sum of volatilities based on daily closing prices for each possible stock pair; (2) allocating each pair to a correlation quintile (ranked fifth) and to a summed volatility quintile; and, (3) randomly selecting 20 twenty pairs out of each of the 25 intersections of correlation and summed volatility quintiles. She accounts for bid-ask frictions by executing all buys (sells) at the ask (bid) and by calculating daily returns at the bid. Using daily bid, ask and closing prices for all stocks included in the S&P 1500 during January 1990 (supporting initial pair trades in January 1991) through December 2014, she finds that: Keep Reading

Do simple moving averages (SMA) commonly used to identify stock market bull and bear regimes work similarly for the spot gold market? To investigate, we consider two market regime indicators: the 200-day SMA and a combination of the 50-day and 200-day SMAs. Because trading days for gold and stocks are sometimes different, we also check a 10-month SMA based on monthly closes. Using daily and monthly spot gold prices and S&P 500 Index levels during January 1973 through April 2016, we find that: Keep Reading

How well does technical trading work for spot currency exchange rates? In their April 2016 paper entitled “Technical Trading: Is it Still Beating the Foreign Exchange Market?”, Po-Hsuan Hsu, Mark Taylor and Zigan Wang 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 focus on average return and Sharpe ratio for measuring rule effectiveness. They use empirical bid-ask spread data as available to estimate costs (averaging 0.045% one way for developed markets and 0.21% one way for emerging markets). They also test whether technical trading effectiveness weakens over time. Using daily U.S. dollar spot exchange rates and associated bid-ask spreads as available for nine developed market currencies and 21 emerging market currencies during January 1971 through mid-September 2015, they find that: Keep Reading

Does adding crash protection based on global market breadth enhance the reliability of dual momentum? In their April 2016 paper entitled “Protective Asset Allocation (PAA): A Simple Momentum-Based Alternative for Term Deposits”, Wouter Keller and Jan Willem Keuning examine a multi-class, dual-momentum portfolio allocation strategy with crash protection based on multi-market breadth. Their principal goal is consistently positive returns, at least 95% (99%) of 1-year rolling returns not below 0% (-5%). Their investment universe is 13 exchange-traded funds (ETF), 12 risky (SPY, QQQ, IWM, VGK, EWJ, EEM, IYR, GSG, GLD, HYG, LQD, TLT) and one safe (IEF). Each month, they:

1. Measure the momentum of each risky ETF as ratio of current price to simple moving average (SMA) of monthly prices over the past 3, 6, 9 or 12 months, minus one.
2. Specify the safe ETF allocation as number of risky assets with non-positive momentum divided by 12 (low crash protection), 9 (medium crash protection) or 6 (high crash protection). For example, if 3 of 12 risky assets have zero or negative momentum, the IEF allocation for high crash protection is 3/6 = 50%.
3. Allocate the balance of the portfolio to the equally weighted 1, 2, 3, 4, 5 or 6 risky assets with the highest positive momentum (reducing the number of risky assets held if not enough have positive momentum).

The interactions of four SMA measurement intervals, three crash protection levels and six risky asset groupings yield 72 combinations. They first identify the optimal combination in-sample during 1971 through 1992 and then test this combination out-of-sample since 1992. Prior to ETF inception dates, they simulate ETF prices based on underlying indexes. They assume constant one-way trading frictions 0.1%, acknowledging that this level may be too low for early years and too high for recent years. They focus on a monthly rebalanced 60% allocation to SPY and 40% allocation to IEF (60/40) as a benchmark. Using monthly simulated/actual ETF total return series during December 1969 through December 2015, they find that: Keep Reading

What is the best way to exploit short-term asset return reversal? In their November 2015 paper entitled “Market Closure and Short-Term Reversal”, Pasquale Della Corte, Robert Kosowski and Tianyu Wang examine four short-term reversal strategies that are each day long (short) assets with below-average (above-average) past returns weighted according to the degree the returns are below (above) average. Portfolio long and short sides are therefore equal in size. Specifically, they assign portfolio weights/accrue portfolio returns based on:

1. Prior close-to-open returns/next open-to-close returns (CO-OC).
2. Prior close-to-close returns/next close-to-close returns (CC-CC).
3. Prior open-to-close returns/next open-to-close returns (OC-OC).
4. Prior open-to-open returns/next open-to-open returns (OO-OO).

For calculating portfolio profitability, they assume a 50% margin requirement. They apply the strategy to each of U.S. stocks, European (French and German) stocks, Japanese stocks, UK stocks, stock index futures, interest rate (bond) futures, commodity futures and currency futures. Using daily opens and closes for stocks since January 1993 and futures since July 1982, all through December 2014, they find that: Keep Reading

Can investors avoid trend trading whipsaws by using Kalman filters to identify trends? In his February 2016 paper entitled “Trend Without Hiccups – A Kalman Filter Approach”, Eric Benhamou investigates the Kalman filter as a tool to smooth (remove the noise from) asset price series in an adaptive way that avoids most of the response lags of moving averages. He defines the Kalman filter as a recursive operation that forecasts the next value in a series based on incomplete and noisy measurements from an assumed distribution of possibilities. He considers four Kalman filter models:

1. Simple model that borrows speed and acceleration concepts from physics with four price inputs to forecast price trajectory.
2. Simple models that calculates a series of local linear trends with five price inputs, in a way very similar to Model 1.
3. More general version of Models 1 and 2 with 10 price inputs measuring both short-term (a few days) and long-term trend contributions.
4. More complex version of Model 3 with 15 price inputs that add effects of oscillations between price extremes over a specified historical interval (14 days).

To test the predictive power of these models, he uses them to identify and trade trends in E-mini S&P 500 futures contract prices. Each trade is only one contract long or short regardless of account value, uses no leverage and includes a $4 round trip commission. Using daily data for E-mini S&P 500 futures from the end of February 2015 through the end of February 2016, he finds that: Keep Reading

Can simple moving average (SMA) rules tell investors when it is prudent to leverage the U.S. stock market? In their March 2016 paper entitled “Leverage for the Long Run – A Systematic Approach to Managing Risk and Magnifying Returns in Stocks”, Michael Gayed and Charles Bilello augment conventional U.S. stock market SMA timing rules by adding leverage while in equities. Specifically, they test a Leverage Rotation Strategy (LRS) comprised of the following rules:

• When the S&P 500 Total Return Index closes above its SMA, hold the index and apply 1.25X, 2X or 3X leverage to magnify returns.
• When the S&P 500 Total Return Index closes below its SMA, switch to U.S. Treasury bills (T-bills) to manage risk.

They focus on a conventional 200-day SMA (SMA200), but include some tests with shorter measurement intervals to gauge robustness. They ignore costs of switching between stocks and T-bills. They apply targeted leverage daily with an assumed 1% annual cost of leverage, approximating current expense ratios for the largest leveraged exchange-traded funds (ETF) that track the S&P 500 Index. Using daily closes of the S&P 500 Total Return Index and T-bill yields during October 1928 through October 2015, they find that: Keep Reading

 “Pervasiveness and Robustness of SMA Effectiveness for Stocks” summarizes research finding that applying a simple moving average (SMA) trading strategy to U.S. stock portfolios produces strong risk-adjusted performance. This strategy is in stocks (cash) when price is above (below) its SMA. Is this finding valid? In his March 2016 paper entitled “Revisiting the Profitability of Market Timing with Moving Averages”, Valeriy Zakamulin challenges the validity of the research. First, he replicates the finding via simulations that incorporate one-month look-ahead bias (by including the last month of SMA calculation intervals as a strategy return). He then corrects strategy return calculations to eliminate this bias. As in the original research, he bases simulations on the following:

• Test data are monthly value-weighted returns of three sets of 10 portfolios from the data library of Kenneth French, each set formed by sorting on market capitalization, book-to-market ratio or momentum.
• Return on cash is the one-month U.S. Treasury bill yield.
• One-way stocks-cash switching cost is 0.5%.
• The sample period is January 1960 through December 2011.

Key performance metrics are net Sharpe ratio and four-factor alpha (adjusting for market, size, book-to-market and momentum factors). Using the specified data and assumptions, he finds that: Keep Reading

“Pervasiveness and Robustness of SMA Effectiveness for Stocks” summarizes research finding that long-term simple moving averages (SMA) pervasively outperform a buy-and-hold approach for U.S. stocks and stock portfolios during 1960-2011 and for seven developed stock markets during 1975-2010. Does this research, which focuses on a 24-month SMA, discover some essential cyclical nature of equity markets? To verify, we test the effectiveness of a 24-month SMA timing strategy versus a buy-and-hold approach for three U.S. stock market series: (1) SPDR S&P 500 (SPY); (2) the underlying S&P 500 Index (augmented with dividend yields estimated from Robert Shiller’s data); and, (3) the Dow Jones Industrial Average (DJIA). The limiting input for the third test is availability of a U.S. Treasury bills (T-bill) yield. The 24-month SMA strategy shifts to stocks (T-bills) when the monthly close crosses above (below) the 24-month SMA. We also test both a 23-month intrinsic momentum strategy, which is in stocks (T-bills) when the lagged 23-month return is positive (negative), and a comparable 10-month SMA strategy. For all timing strategies, we assume an investor can slightly anticipate signals and execute trades at the same close. Using monthly returns for SPY since January 1993 (dividend-adjusted), the S&P 500 Index with dividend yields from Shiller since January 1950 and DJIA since January 1932, along with contemporaneous monthly 3-month T-bill yields, all through January 2016, we find that: Keep Reading

A subscriber suggested testing of the Guardian Indicator, “a proprietary new market-strength indicator designed to enhance risk-adjusted investment returns by identifying long-term directional changes in the stock market.” This indicator tabulates Guard Score (GS) “votes” by U.S. equity sectors to predict the trend of the overall U.S. stock market. Per the paper “Introducing Guardian Indicator: Market Timing Based on Momentum and Volatility”, flagged by the subscriber, GS is the greater of two contributing indicators:

• Price momentum indicator (PI), the ratio of the 50-day simple moving average (SMA) to the 200-day SMA of asset/index level.
• Volatility regime indicator (VI), the ratio of the 1250-day SMA to the 250-day SMA of downside deviation, calculated as the square root of the sum of squared negative daily returns over the past 90 trading days divided by 90.

If GS is greater than one, the trend is bullish. The paper applies GS to time the S&P 500 Index during 1957-2014 (apparently without dividends). Here we replicate the GS series for the S&P 500 Index (excluding dividends) and use it to time the index. In calculating returns, however, we account for S&P 500 dividends by each month allocating the annual dividend yield from Robert Shiller’s data to the days of that month (dividing by 252). We assume cash earns the 3-month U.S. Treasury bill (T-bill) yield. We invest in the S&P 500 Index (T-bills) when prior-day GS for the index is greater than (less than or equal to) one. We focus on compound annual growth rate (CAGR) and maximum drawdown (MaxDD) as performance measures. We use total return from buying and holding the S&P 500 Index (B&H) as a benchmark. Using daily S&P 500 Index level and monthly S&P 500 dividend yield and T-bill yield during January 1950 through December 2015 (allowing first calculation of VI in May 1955), we find that: Keep Reading