Momentum Investing

Which economic and market variables are most effective in predicting U.S. stock market returns? In his October 2018 paper entitled “Forecasting US Stock Returns”, David McMillan tests 10-year rolling and recursive (inception-to-date) one-quarter-ahead forecasts of S&P 500 Index capital gains and total returns using 18 economic and market variables, as follows: dividend-price ratio; price-earnings ratio; cyclically adjusted price-earnings ratio; payout ratio; Fed model; size premium; value premium; momentum premium; quarterly change in GDP, consumption, investment and CPI; 10-year Treasury note yield minus 3-month Treasury bill yield (term structure); Tobin’s q-ratio; purchasing managers index (PMI); equity allocation; federal government consumption and investment; and, a short moving average. He tests individual variables, four multivariate combinations and and six equal-weighted combinations of individual variable forecasts. He employs both conventional linear statistics and non-linear economic measures of accuracy based on sign and magnitude of forecast errors. He uses the historical mean return as a forecast benchmark. Using quarterly S&P 500 Index returns and data for the above-listed variables during January 1960 through February 2017, he finds that: Keep Reading

How does a large sample of stock return anomalies fare in recent replication testing? In their October 2018 paper entitled “Replicating Anomalies”, Kewei Hou, Chen Xue and Lu Zhang attempt to replicate 452 published U.S. stock return anomalies, including 57, 69, 38, 79, 103, and 106 anomalies 57 momentum, 69 value-growth, 38 investment, 79 profitability, 103 intangibles and 106 trading frictions (trading volume, liquidity, market microstructure) anomalies. Compared to the original papers, they use the same sample populations, original (as early as January 1967) and extended (through 2016) sample periods and similar methods/variable definitions. They test limiting influence of microcaps (stocks in the lowest 20% of market capitalizations) by using NYSE (not NYSE-Amex-NASDAQ) size breakpoints and value-weighted returns. They consider an anomaly replication successful if average high-minus-low tenth (decile) return is significant at the 5% level, translating to t-statistic at least 1.96 for pure standalone tests and at least 2.78 assuming multiple testing (accounting for aggregate data snooping bias). Using required anomaly data and monthly returns for U.S. non-financial stocks during January 1967 through December 2016, they find that:

Keep Reading

Quantitative investing involves disciplined rule-based approaches to help investors structure optimal portfolios that balance return and risk. How has such investing evolved? In their June 2018 paper entitled “The Current State of Quantitative Equity Investing”, Ying Becker and Marc Reinganum summarize key developments in the history of quantitative equity investing. Based on the body of research, they conclude that: Keep Reading

Brian Livingston introduces his 2018 book, Muscular Portfolios: The Investing Revolution for Superior Returns with Lower Risk, as follows: “What we laughingly call the financial ‘services’ industry is a cesspool filled with sharks intent on siphoning your money away and making it their own. The good news is that it is absolutely possible to grow your savings with no fear of financial sharks or stock market crashes. In the past few years, we’ve seen an explosion of low-cost index funds, along with serious mathematical breakthroughs in how to combine these funds into low-risk portfolios. …This book shows you how.  …You can start with just a little money and make it grow.” Based on research from multiple sources and extensions of that research, he concludes that: Keep Reading

A subscriber asked whether applying a filter that restricts monthly asset selections of the “Simple Asset Class ETF Momentum Strategy” (SACEMS) to those currently at an intermediate-term high improves performance. This strategy each month reforms a portfolio of winners from the following universe based on total return over a specified lookback interval:

PowerShares DB Commodity Index Tracking (DBC)
iShares MSCI Emerging Markets Index (EEM)
iShares MSCI EAFE Index (EFA)
SPDR Gold Shares (GLD)
iShares Russell 2000 Index (IWM)
SPDR S&P 500 (SPY)
iShares Barclays 20+ Year Treasury Bond (TLT)
Vanguard REIT ETF (VNQ)
3-month Treasury bills (Cash)

To investigate, we focus on the equally weighted (EW) Top 3 SACEMS portfolio and replace any selection not at an intermediate-term high with Cash. We define intermediate-term high based on monthly closes over a specified past interval ranging from one month to six months. We consider all gross performance metrics used for base SACEMS. Using monthly dividend adjusted closing prices for the asset class proxies and the yield for Cash over the period February 2006 (the earliest all ETFs are available) through July 2018 (150 months), we find that: Keep Reading

From a purely statistical perspective, how many factors are optimal for explaining both time series and cross-sectional variations in stock anomaly/stock returns, and how do these statistical factors relate to stock/firm characteristics? In their July 2018 paper entitled “Factors That Fit the Time Series and Cross-Section of Stock Returns”, Martin Lettau and Markus Pelger search for the optimal set of equity factors via a generalized Principal Component Analysis (PCA) that includes a penalty on return prediction errors returns. They apply this approach to three datasets:

1. Monthly returns during July 1963 through December 2017 for two sets of 25 portfolios formed by double sorting into fifths (quintiles) first on size and then on either accruals or short-term reversal.
2. Monthly returns during July 1963 through December 2017 for 370 portfolios formed by sorting into tenths (deciles) for each of 37 stock/firm characteristics.
3. Monthly excess returns for 270 individual stocks that are at some time components of the S&P 500 Index during January 1972 through December 2014.

They compare performance of their generalized PCA to that of conventional PCA. Using the specified datasets, they find that: Keep Reading

Are there any gold trading strategies that reliably beat buy-and-hold? In their April 2018 paper entitled “Investing in the Gold Market: Market Timing or Buy-and-Hold?”, Viktoria-Sophie Bartsch, Dirk Baur, Hubert Dichtl and Wolfgang Drobetz test 4,095 seasonal, 18 technical, and 15 fundamental timing strategies for spot gold and gold futures. These strategies switch at the end of each month as signaled between spot gold or gold futures and U.S. Treasury bills (T-bill) as the risk-free asset. They assume trading frictions of 0.2% of value traded. To control for data snooping bias, they apply the superior predictive ability multiple testing framework with step-wise extensions. Using monthly spot gold and gold futures prices and T-bill yield during December 1979 through December 2015, with out-of-sample tests commencing January 1990, they find that:

Keep Reading

The many factor-based indexes and exchange-traded funds (ETFs) that track them now available enable investors to construct multi-factor portfolios piecemeal. Is such piecemeal construction suboptimal? In their July 2018 paper entitled “The Characteristics of Factor Investing”, David Blitz and Milan Vidojevic apply a multi-factor expected return linear regression model to explore behaviors of long-only factor portfolios. They consider six factors: value-weighted market, size, book-to-market ratio, momentum, operating profitability and investment(change in assets). Their model generates expected returns for each stock each month, and further aggregates individual stock expectations into factor-portfolio expectations holding all other factors constant. They use the model to assess performance differences between a group of long-only single-factor portfolios and an integrated multi-factor portfolio of stocks based on combined rankings across factors. The focus on gross monthly excess (relative to the 10-year U.S. Treasury note yield) returns as a performance metric. Using data for a broad sample of U.S. common stocks among the top 80% of NYSE market capitalizations and priced at least $1 during June 1963 through December 2017, they find that: Keep Reading

A betting against beta (BAB) portfolio is long low-beta assets and short high-beta assets, with each side leveraged to a beta of one. Do strong past stock market returns (when investors tend to overweight high-beta stocks) predict an increase in BAB returns? In his June 2018 paper entitled “Time-Varying Leverage Demand and Predictability of Betting-Against-Beta”, Esben Hedegaard tests the prediction that BAB performs better in times and in countries after high past stock market returns in three ways: (1) regression of BAB returns versus past market returns; (2) sorts of BAB returns into fifths (quintiles) based on past market returns; and, (3) a timing strategy that is long BAB half the time and short BAB half the time based on detrended inception-to-date past market returns, scaled to 10% annualized volatility for comparability. Using daily and monthly data, including monthly BAB returns, for U.S. common stocks and the U.S. stock market since 1931 and for 23 other countries from as early as 1988, all through January 2018, he finds that: Keep Reading

A subscriber asked whether some different momentum metric might improve performance of the “Simple Asset Class ETF Momentum Strategy” (SACEMS), which each month reforms a portfolio of winners from the following universe based on total return over a specified lookback interval:

PowerShares DB Commodity Index Tracking (DBC)
iShares MSCI Emerging Markets Index (EEM)
iShares MSCI EAFE Index (EFA)
SPDR Gold Shares (GLD)
iShares Russell 2000 Index (IWM)
SPDR S&P 500 (SPY)
iShares Barclays 20+ Year Treasury Bond (TLT)
Vanguard REIT ETF (VNQ)
3-month Treasury bills (Cash)

To investigate, we compare performances of the following alternative monthly momentum metrics to that of the original baseline metric:

• Average monthly total returns over the lookback interval.
• Slope of the dividend-adjusted price series over the lookback interval.
• Sharpe ratio of the monthly total return series over the lookback interval (using Cash return as the risk-free rate, and setting the Sharpe ratio of Cash at zero).

We focus on the equally weighted (EW) Top 3 SACEMS portfolio. We consider all the performance metrics used for the baseline, with emphasis on compound annual growth rates (CAGR) and maximum drawdowns (MaxDD). Using monthly dividend adjusted closing prices for the asset class proxies and the yield for Cash over the period February 2006 (the earliest all ETFs are available) through May 2018 (148 months), we find that: Keep Reading