Momentum Investing

How sensitive is the performance of the “Simple Sector ETF Momentum Strategy” to selecting ranks other than winners and to choosing a momentum ranking interval other than six months? This strategy each month ranks the following nine sector exchange-traded funds (ETF) on past return and rotates to the strongest sector:

Materials Select Sector SPDR (XLB)
Energy Select Sector SPDR (XLE)
Financial Select Sector SPDR (XLF)
Industrial Select Sector SPDR (XLI)
Technology Select Sector SPDR (XLK)
Consumer Staples Select Sector SPDR (XLP)
Utilities Select Sector SPDR (XLU)
Health Care Select Sector SPDR (XLV)
Consumer Discretionary Select SPDR (XLY)

Available data are so limited that sensitivity test results may mislead. With that reservation, we perform two robustness/sensitivity tests: (1) comparison of returns for all nine ranks of winner through loser based on a ranking interval of six months and a holding interval of one month (6-1); and, (2) comparison of winner returns for ranking intervals ranging from one to 12 months (1-1 through 12-1) and for a six-month lagged six-month ranking interval (12:7-1) per “Isolating the Decisive Momentum (Echo?)”, all with one-month holding intervals. Using monthly adjusted closing prices for the sector ETFs and SPDR S&P 500 (SPY) over the period December 1998 through December 2015 (205 months), we find that: Keep Reading

Do simple momentum trading strategies applied to major U.S. stock market sectors outperform reasonable benchmarks? To investigate, we apply three simple momentum strategies to the nine sector exchange-traded funds (ETF) defined by the Select Sector Standard & Poor’s Depository Receipts (SPDR):

Materials Select Sector SPDR (XLB)
Energy Select Sector SPDR (XLE)
Financial Select Sector SPDR (XLF)
Industrial Select Sector SPDR (XLI)
Technology Select Sector SPDR (XLK)
Consumer Staples Select Sector SPDR (XLP)
Utilities Select Sector SPDR (XLU)
Health Care Select Sector SPDR (XLV)
Consumer Discretionary Select SPDR (XLY)

The three strategies are: (1) allocate all funds at the end of each month to the sector ETF with the highest total return over the past six months (6-1); (2) allocate all funds at the end of each month to the sector ETF with the highest total return over the six months ending the prior month (6-1;1), hypothesizing that the skip-month avoids short-term reversals; and, (3) more cautiously, allocate all funds at the end of each month either to the sector ETF with the highest total return over the past six months or to cash depending on whether the S&P 500 Index is above or below its 10-month simple moving average (6-1;SMA10). A six-month ranking period is intuitively large enough to gauge sector momentum but small enough to react to changes in business conditions that might favor one sector over others. Using monthly dividend-adjusted closing prices for the sector ETFs, the S&P 500 index, 3-month Treasury bills (T-bills) and S&P Depository Receipts (SPY) over the period December 1998 through December 2015 (205 months), we find that: Keep Reading

Does asset allocation based on both trend following via a simple moving average (SMA) and return momentum work well? In the July 2015 update of their paper entitled “The Trend is Our Friend: Risk Parity, Momentum and Trend Following in Global Asset Allocation”, Andrew Clare, James Seaton, Peter Smith and Stephen Thomas examine the effectiveness of trend following based on SMAs and momentum screens in forming portfolios across and within asset classes. They consider five asset classes: developed equity markets (24 component country indexes); emerging equity markets (16 component country indexes); bonds (19 component country indexes); commodities (23 component commodity indexes); and, real estate (13 country REIT indexes). They compare equal weight and risk parity (proportional to inverse 12-month volatility) strategic allocations. They define trend following as buying (selling) an asset when its price moves above (below) a moving average of 6, 8, 10 or 12 months. They consider both simple momentum (12-month lagged total return) and volatility-adjusted momentum (dividing by standard deviation of monthly returns over the same 12 months) for momentum screens. They ignore trading frictions, exclude shorting and assume monthly trend/momentum calculations and associated trade executions are coincident. Using monthly total returns in U.S. dollars for the five broad value-weighted asset class indexes and for the 95 components of these indexes during January 1993 through March 2015, along with contemporaneous 3-month Treasury bill yields as the return on cash, they find that: Keep Reading

Does the information in short, intermediate and long stock price trends combined by relating multiple simple moving averages (SMA) to future returns usefully predict stock returns? In the September 2015 update of their paper entitled “A Trend Factor: Any Economic Gains from Using Information over Investment Horizons?”, Yufeng Han and Guofu Zhou examine a trend factor that simultaneously captures short, intermediate and long stock price trends. Specifically, at the end of each month for each sampled stock, they:

1. Calculate SMAs over the past 3, 5, 10, 20, 50, 100, 200, 400, 600, 800 and 1,000 trading days.
2. Normalize SMAs by dividing by the final close.
3. Regress monthly SMAs against next-month stock returns to estimate historical linear coefficients for all SMAs.
4. Predict the return for the stock next month based on average SMA coefficients for the past 12 months applied to the most recent set of SMAs.

They define the trend factor as the average monthly gross return for a portfolio that is each month long (short) the equally weighted fifth (quintile) of stocks with the highest (lowest) expected returns. Using daily prices and associated stock/firm characteristics for a broad sample of U.S. common stocks during January 1926 through December 2014, they find that: Keep Reading

Is it possible to test factor models of stock returns directly on individual stocks rather than on portfolios of stocks sorted per preconceived notions of factor importance. In their November 2015 paper entitled “Tests of Alternative Asset Pricing Models Using Individual Security Returns and a New Multivariate F-Test”, Shafiqur Rahman, Matthew Schneider and Gary Antonacci apply a statistical method that allows testing of equity factor models directly on individual stocks. Results are therefore free from the information loss and data snooping bias associated with sorting stocks based on some factor into portfolios. They test several recently proposed multi-factor models based on five or six of market, size, value (different definitions), momentum, liquidity (based on turnover), profitability and investment factors. They compare alternative models via 100,000 Monte Carlo simulations each in terms of ability to eliminate average alpha and appraisal ratio (absolute alpha divided by residual variance) across individual stocks. Using monthly returns and stock/firm characteristics for the 407 Russell 3000 Index stocks with no missing monthly returns during January 1990 through December 2014 (300 months), they find that: Keep Reading

How do the behaviors of time-series (absolute) and cross-sectional (relative) carry, momentum and value strategies differ? In the November 2015 version of their paper entitled “Dissecting Investment Strategies in the Cross Section and Time Series”, Jamil Baz, Nicolas Granger, Campbell Harvey, Nicolas Le Roux and Sandy Rattray explore time-series and cross-sectional carry, momentum and value strategies as applied to multiple asset classes. They adapt to each asset class the following general definitions:

• Carry – buy (sell) futures on assets for which the forward price is lower (higher) than the spot price.
• Momentum – buy (sell) assets that have outperformed (underperformed) over the past 6-12 months.
• Value – buy (sell) assets for which market price is lower (higher) than estimated fundamental price.

For cross-sectional portfolios, they rank assets within each class-strategy and form portfolios that are long (short) the equally weighted six assets with the highest (lowest) expected returns, rebalanced daily except for currency carry and value trades. For time-series portfolios, they take an equal long (short) position in each asset within a class-strategy according to whether its expected return is positive (negative). When combining strategies within an asset class, they use equal weighting. When combining across asset classes, they scale each class-strategy portfolio to a 15% annualized volatility target. Using daily contract closing bid-ask midpoints for 26 equity futures, 14 interest rate swaps, 31 currency exchange rates and 16 commodity futures during January 1990 through April 2015, they find that: Keep Reading

A subscriber requested evaluation of Jay’s Pure Momentum Sector Fund System, specified by originator Jay Kaeppel as follows:

• At the end of the first month, assign 20% weight to the five of the 40 Fidelity Select Sector funds (excluding Select Gold, FSAGX) with the largest positive returns over the previous 240 trading days.
• At the end of each subsequent month, sell any positions that drop out of the top five and reallocate proceeds equally to their replacements.
• If for any month fewer than five funds have positive returns, leave unpopulated positions in cash.

This system involves both relative momentum (picking past winners) and absolute or intrinsic momentum (requiring positive past returns). The author states that the publication year for the system is 2001, so we start with 2002 for a test free of data snooping. We accept annual returns for 2002 through (partial) 2015 as reported by the author . We consider two simple benchmarks: (1) buy and hold SPDR S&P 500 (SPY); and, (2) hold SPY when it is above its 10-month simple moving average and 3-month U.S. Treasury bills (T-bills, a proxy for cash) otherwise (SPY-SMA10). The second benchmark is a simple, widely used market timing rule that helps decide whether Jay’s Pure Momentum Sector Fund System outperforms the market because of sector rotation (relative momentum) or market timing (absolute momentum). Using annual returns for Jay’s Pure Momentum System, monthly dividend-adjusted prices and annual returns for SPY and monthly T-bill yields during 2002 through mid-September 2015 (nearly 14 years), we find that: Keep Reading

Does “meaningful” short-term stock return momentum predict exploitable short-term price trends? In their October 2015 paper entitled “News Momentum”, Hao Jiang, Sophia Li  and Hao Wang combine time-stamped firm news with high-frequency (15-minute) stock returns to identify stocks exhibiting news-driven momentum. Their news feed is the stream of unique items (no repeated stories) delivered in near real time by RavenPack. News-driven momentum derives from high-frequency returns that coincide with real-time news, arguably capturing the reactions of the most attentive investors. To test exploitability of news momentum, they each day form an equally weighted hedge portfolio that is long (short) the tenth, or decile, of stocks with the highest (lowest) prior-day news-driven momentum and hold for five trading days. On any given day, they calculate strategy return as the average return of the current five overlapping portfolios. Using intraday stock price/quote data, associated firm news and other stock/firm data for a broad sample of U.S. common stocks during March 2000 through October 2012, they find that: Keep Reading

Are trend following (intrinsic or time series momentum) and risk parity complementary multi-class portfolio construction approaches? In his October 2015 paper entitled “Trend-Following, Risk-Parity and the Influence of Correlations”, Nick Baltas compares performances of inverse volatility weighting and risk parity weighting as adapted to a long-short trend following strategy. Unlike volatility weighting, risk parity weighting incorporates asset return correlations, assigning higher (lower) weights to assets with lower (higher) average pairwise correlations with other assets. For both weighting schemes, portfolios are each month long (short) assets with positive (negative) past 12-month returns. Monthly inverse volatility weights derive from actual daily asset return volatilities over the past 90 trading days. Monthly risk parity weights derive from actual daily asset return volatilities and correlations over the past 90 trading days. Both weighting schemes target 10% portfolio volatility by each month applying overall leverage based on actual annualized volatility of an unleveraged trend following portfolio over the past 60 trading days divided by 10%. Using daily closing prices for the most liquid contract for each of 35 (6 energy, 10 commodity, 6 government bond, 6 currency exchange rate and 7 equity index) futures contract series as available during January 1987 through December 2013, he finds that: Keep Reading

Is there a simple way to improve the performance of conventional asset class target allocations by rotating to strength within classes based on Relative Strength Index (RSI)? In his September 2015 paper entitled “Momentum Investing and Asset Allocation”, Drew Knowles seeks to improve the performance of baseline asset class (equity, fixed income, hedge fund) allocations via dynamic intra-class rotation to strength based on RSI. His principal passive benchmark (50/30/20) allocates 50% to equities (S&P 500 Total Return Index), 30% to fixed income (Barclays U.S. Aggregate Index) and 20% to hedge funds (HFRI Fund Weighted Composite), apparently rebalanced annually. For dynamic rotation, he replaces the broad equity, fixed income and hedge fund indexes with, respectively, the apparently equally weighted Top 5 (of 10) S&P 500 sector indexes, Top 5 (of seven) fixed income style indexes and Top 5 (of eight) hedge fund style indexes based on 12-month RSI. He breaks ties in RSI by picking the index with higher return per unit of risk (compound annual growth rate divided by standard deviation of returns) over the same 12 months. Within each asset class, he tests four Top 5 reformation frequencies: annual, semi-annual, quarterly or monthly. He ignores rebalancing/reformation frictions and tax implications of trading. Using monthly data for the selected broad and sector/style indexes during 1991 through 2014, he finds that: Keep Reading