Objective research to aid investing decisions
Menu
Value Allocations for Apr 2019 (Final)
Cash TLT LQD SPY
Momentum Allocations for Apr 2019 (Final)
1st ETF 2nd ETF 3rd ETF

Investing Research Articles

Online, Real-time Test of AI Stock Picking?

Will equity funds “managed” by artificial intelligence (AI) outperform human investors? To investigate, we consider the performance of AI Powered Equity ETF (AIEQ), which “seeks to provide investment results that exceed broad U.S. Equity benchmark indices at equivalent levels of volatility.” More specifically, offeror EquBot: “…leverages IBM’s Watson AI to conduct an objective, fundamental analysis of U.S.-listed common stocks and real estate investment trusts…based on up to ten years of historical data and apply that analysis to recent economic and news data. Each day, the EquBot Model ranks each company based on the probability of the company benefiting from current economic conditions, trends, and world events and identifies approximately 30 to 70 companies with the greatest potential over the next twelve months for appreciation and their corresponding weights, while maintaining volatility…comparable to the broader U.S. equity market. The Fund may invest in the securities of companies of any market capitalization. The EquBot model recommends a weight for each company based on its potential for appreciation and correlation to the other companies in the Fund’s portfolio. The EquBot model limits the weight of any individual company to 10%.” We use SPDR S&P 500 (SPY) as a simple benchmark for AIEQ performance. Using daily dividend-adjusted closes of AIEQ and SPY from AIEQ inception (October 18, 2017) through December 2018, we find that: Keep Reading

Combining Fundamental Analysis and Portfolio Optimization

Can stock return forecasts from fundamental analysis make conventional mean-variance stock portfolio optimization work? In their December 2018 paper entitled “Optimized Fundamental Portfolios”, Matthew Lyle and Teri Yohn construct a portfolio that combines fundamentals-based stock return forecasts and mean-variance optimization and then compare results with portfolios from each employed separately. To suppress implementation costs, they focus on long-only portfolios reformed quarterly. Their fundamentals return forecasting model uses cross-sectionally normalized versions of book-to-market ratio, return on equity, change in net operating assets divided by book value and change in financial assets divided by book value. They update fundamental variables quarterly at the end of the reporting month. They generate stock return forecasts via a complicated multivariate regression of cross-sectionally normalized versions of the variables based on five years of rolling historical data. They then form a portfolio of the tenth (decile) of stocks with the highest expected returns, either value-weighted or equal-weighted. They consider several portfolio optimization methods, including minimum variance (requiring no return forecasts); mean-variance optimization with target expected return; and, Sharpe ratio maximization. Their combined approach employs fundamental stock return forecasts as inputs to those portfolio optimization methods that require returns. They use data from 1991-1995 to generate initial model inputs and 1996-2015 for out-of-sample testing. Using end-of-month data for a broad but groomed sample of U.S. common stocks with at least three years of historical data during January 1991 through December 2015, they find that:

Keep Reading

Trend Following: Momentum or Moving Average?

Are moving averages or intrinsic (time series) momentum theoretically better for following trends in asset prices? In their November 2018 paper entitled “Trend Following with Momentum Versus Moving Average: A Tale of Differences”, Valeriy Zakamulin and Javier Giner compare from a theoretical perspective effectiveness of four popular trend following rules:

  1. Intrinsic Momentum – buy (sell) when the closing price at the end of a specified lookback interval is greater (less) than the closing price at the beginning of the lookback interval.
  2. Simple Moving Average – buy (sell) when the closing price at the end of a specified lookback interval is greater (less) than the equally weighted average closing price during the lookback interval.
  3. Linear Moving Average – buy (sell) when the closing price at the end of a specified lookback interval is greater (less) than the linearly weighted (weights linearly increasing to the most recent) average closing price during the lookback interval.
  4. Exponential Moving Average – buy (sell) when the closing price at the end of a specified lookback interval is greater (less) than the exponentially weighted (weights exponentially increasing to the most recent) average closing price during the lookback interval.

They transform these price rules into return-based versions and create a trend model as an autoregressive return process. They then explore interactions of the trading rules with the trend model. Based on this theoretical approach, they conclude that: Keep Reading

Momentum and Bubble Stocks

Do “bubble” stocks (those with high shorting demand and small borrowing supply) exhibit unconventional momentum behaviors? In their December 2018 paper entitled “Overconfidence, Information Diffusion, and Mispricing Persistence”, Kent Daniel, Alexander Klos and Simon Rottke examine how momentum effects for bubble stocks differ from conventional momentum effects. They each month sort stocks into groups independently as follows:

  1. Momentum winners (losers) are the 30% of stocks with the highest (lowest) returns from one year ago to one month ago, incorporating a skip-month.
  2. Stocks with high (low) shorting demand are those with the top (bottom) 30% of short interest ratios.
  3. Stocks with small (large) borrowing supply are those with the top (bottom) 30% of institutional ownerships.

They then use intersections of these groups to reform 27 value-weighted portfolios. Bubble (constrained) stocks are those in the intersection of high shorting demand and low institutional ownership, including both momentum winners and losers. For purity, they further split bubble losers into those that were or were not also bubble winners within the past five years. Using monthly and daily returns, market capitalizations and trading volumes for a broad sample of U.S. common stocks, monthly short interest ratios and quarterly institutional ownership data from SEC Form 13F filings during July 1988 through June 2018, they find that: Keep Reading

Weekly Summary of Research Findings: 1/7/19 – 1/11/19

Below is a weekly summary of our research findings for 1/7/19 through 1/11/19. These summaries give you a quick snapshot of our content the past week so that you can quickly decide what’s relevant to your investing needs.

Subscribers: To receive these weekly digests via email, click here to sign up for our mailing list. Keep Reading

Robustness of SACEMS Based on Sharpe Ratio

Subscribers have asked whether risk-adjusted returns might work better than raw returns for ranking Simple Asset Class ETF Momentum Strategy (SACEMS) assets. In fact, “Alternative Momentum Metrics for SACEMS?” supports belief that Sharpe ratio beats raw returns. Is this finding strong enough to justify changing the strategy, which each month selects the best performers over a specified lookback interval from among the following eight asset class exchange-traded funds (ETF), plus cash:

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 update the basic comparison and conduct three robustness tests:

  1. Does Sharpe ratio beat raw returns consistently across Top 1, equally weighted (EW) Top 2, EW Top 3 and EW Top 4 portfolios, and the 50%-50% SACEMS EW Top 3-Simple Asset Class ETF Value Strategy (SACEVS) Best Value portfolio?
  2. Does Sharpe ratio beat raw returns consistently across different lookback intervals?
  3. For multi-asset portfolios, does weighting by Sharp ratio rank beat equal weighting? In other words, do future returns behave systematically across ranks?

To calculate Sharpe ratios, we each month for each asset subtract the risk-free rate (Cash yield) from raw monthly total returns to generate monthly total excess returns over a specified lookback interval. We then calculate Sharpe ratio as average monthly excess return divided by standard deviation of monthly excess returns over the lookback interval. We set Sharpe ratio for Cash at zero (though it is actually zero divided by zero). Using monthly dividend-adjusted closing prices for asset class proxies and the yield for Cash during February 2006 (when all ETFs are first available) through December 2018, we find that: Keep Reading

Pump-and-Dump Participation/Losses

A “pump-and-dump” scheme promoter: (1) builds a position in a stock (often a thinly traded penny stock); (2) gooses its price by spreading misleading information; and, (3) liquidates the position once the stock reaches. Who responds to such schemes and what are their returns? In the December 2018 revision of their paper entitled “Who Falls Prey to the Wolf of Wall Street? Investor Participation in Market Manipulation”, Christian Leuz, Steffen Meyer, Maximilian Muhn, Eugene Soltes and Andreas Hackethal investigate pump-and-dump scheme participation rate, purchase size/returns and participant characteristics. Specifically, they explore the intersection of 421 such schemes (both from the responsible German regulatory agency and hand-selected) and trading records/demographics for 113,000 randomly selected individual investors from a major German bank. Using the specified data spanning 2002 through 2015, they find that:

Keep Reading

Aggregate Patent Value as Stock Return Predictor

Is value of a firm’s patents a reliable predictor of its stock returns? In their November 2018 paper entitled “Patent-to-Market Premium”, Jiaping Qiu, Kevin Tseng and Chao Zhang investigate firm patent-to-market (PTM) ratio (percentage of market value attributable to patents) as a predictor of stock returns. They specify PTM ratio for each firm as follows:

  1. Measure stock reaction to each patent grant date.
  2. Depreciate each patent since grant date via an inventory depreciation method.
  3. Estimate cumulative market value of all patents held by adding current depreciated values.
  4. Divide cumulative patent value by firm market value.

They then at the end of June each year reform a hedge portfolio that is long (short) the tenth, or decile, of stocks with the highest (lowest) PTM ratios. Using market and lagged accounting data for a broad sample of U.S. common stocks/firms and intersecting patent data (5,475 distinct firms) during 1965 through 2010, they find that: Keep Reading

Does Active Stock Factor Timing/Tilting Work?

Does active stock factor exposure management boost overall portfolio performance? In their November 2018 paper entitled “Optimal Timing and Tilting of Equity Factors”, Hubert Dichtl, Wolfgang Drobetz, Harald Lohre, Carsten Rother and Patrick Vosskamp explore benefits for global stock portfolios of two types of active factor allocation:

  1. Factor timing – exploit factor premium time series predictability based on economic indicators and factor-specific technical indicators.
  2. Factor tilting – exploit cross-sectional (relative) attractiveness of factor premiums.

They consider 20 factors spanning value, momentum, quality and size. For each factor each month, they reform a hedge portfolio that is long (short) the equal-weighted fifth, or quintile, of stocks with the highest (lowest) expected returns for that factor. For implementation of factor timing, they consider: 14 economic indicators standardized by subtracting respective past averages and dividing by standard deviations; and, 16 technical indicators related to time series momentum, moving averages and volatilities. They suppress redundancy and noise in these indicators via principal component analysis separately for economic and technical groups, focusing on the first principal component of each group. They translate any predictive power embedded in principal components into optimal factor portfolio weights using augmented mean-variance optimization. For implementation of factor tilting, they overweight (underweight) factors that are relatively attractive (unattractive) based on valuations of factor top and bottom quintile stocks, top-bottom quintile factor variable spreads, prior-month factor returns (momentum) and volatilities of past monthly factor returns. Their benchmark portfolio is the equal-weighted combination of all factor hedge portfolios. For all portfolios, they assume: monthly portfolio reformation costs of 0.75% (1.15%) of turnover value for the long (short) side; and, annual 0.96% cost for an equity swap to ensure a balanced portfolio of factor portfolios. For monthly factor timing and tilting portfolios only, they assume an additional cost of 0.20% of associated turnover. Using monthly data for a broad sample of global stocks from major equity indexes and for specified economic indicators during January 1997 through December 2016 (4,500 stocks at the beginning and 5,000 stocks at the end), they find that: Keep Reading

Momentum and Stock Return Dispersion

Is stock price momentum an imperfect proxy for sensitivity of individual stocks to past dispersion of returns across stocks (zeta risk, or return dispersion)? In their November 2018 paper entitled “Market Risk and the Momentum Mystery”, James Kolari and Wei Liu investigate relationships between momentum and return dispersion as predictors of individual U.S. stock returns. They employ both portfolio comparisons and regression tests. For the former, their momentum portfolio is long (short) the equally weighted top (bottom) tenth, or decile, of stocks ranked on past 12-month minus one skip-week returns, reformed monthly. Their main return dispersion portfolio is long (short) the equally weighted decile of stocks with the most positive (negative) sensitivities to the dispersion of all individual daily stock returns over the past 12 months minus one skip-week, reformed monthly. Using daily and monthly returns for a broad sample of U.S. stocks priced over $5 since January 1964, and contemporaneous 1-month U.S. Treasury bill yields and monthly returns of selected stock return model factors since January 1965, all through December 2017, they find that:

Keep Reading

Daily Email Updates
Login
Research Categories
Recent Research
Popular Posts