Equity Premium

Are concentrated stock portfolios inherently disadvantaged by lack of diversification? In his June 2023 paper entitled “Underperformance of Concentrated Stock Positions”, Antti Petajisto analyzes rolling future returns for individual U.S. stocks relative to the broad U.S. stock market (market-adjusted) as a way to assess implications of concentrated stock portfolios. He focuses on median return as most representative of investor experience. He considers monthly rolling investment horizons of five, 10 and 20 years because concentrated stock positions are typically long-term holdings. He looks also at the relationship between 5-year past returns and future returns for individual stocks. Using monthly returns for individual U.S. common stocks from an evolving sample similar to the Russell 3000 (no microcaps) and for the overall capitalization-weighted U.S. stock market during January 1926 through December 2022, he finds that:

Keep Reading

Are new machine learning portfolio strategies practically better than old stock factor ways? In their August 2023 paper entitled “Predicting Returns with Machine Learning Across Horizons, Firms Size, and Time”, Nusret Cakici, Christian Fieberg, Daniel Metko and Adam Zaremba examine the ability of various machine learning models to predict stock returns for: (1) monthly and annual return forecast horizons; (2) three ranges of firm size; and, (3) two subperiods. They apply eight machine learning models (including simple and penalized linear regressions, dimension reduction techniques, regression trees and neural networks) to 153 firm/stock characteristics following approaches typical in the finance literature. For each model, they employ rolling 11-year intervals, with:

• Model training using the first seven years.
• Model validation using the next three years.
• Out-of-sample testing the last year using hedge portfolios that are long (short) the value-weighted fifth, or quintile, of stocks with the highest (lowest) predicted returns, reformed either monthly or annually depending forecast horizon.

They focus on gross 6-factor (market, size, book-to-market, profitability, investment, momentum) alpha to assess machine learning effectiveness. Using data for the selected 153 firm/stock characteristics and associated stock returns, measured monthly, for all listed U.S. stocks during January 1972 through December 2020, they find that: Keep Reading

The conventional approach to measuring equity factor returns is via hedge portfolios that are long (short) the equal-weighted or value-weighted extreme highest (lowest) fifth or tenth of stocks sorted by some firm/stock characteristic. Is there a better way? In their August 2023 paper entitled “Power Sorting”, Anastasios Kagkadis, Harald Lohre, Ingmar Nolte, Sandra Nolte and Nikolaos Vasilas construct equity factor portfolios based on power sorting, which: (1) models the firm characteristic-future stock return relationship using a power series; and, (2) uses the power series to set factor portfolio weights. This approach requires no arbitrary quantile break points, instead allocating some weight to all stocks and tilting toward/away stocks with extreme characteristics as a compromise between conventional sorts and machine learning methods. Power sorting employs separate parameters for the long and short sides of the factor portfolio. Higher parameter values generate portfolios that concentrate more in stocks with characteristic extremes, while lower values spread weights more evenly across stocks. Differences between the two parameters allow differently weighted long and short sides of a factor portfolio. Additionally, they set an upper limit on the allocation to any one stock (2% in their main analysis) to ensure factor portfolio diversification. Using monthly factor data and associated stock returns for 85 widely accepted factor characteristics during March 1980 through December 2021, they find that:

Keep Reading

Should investors trust artificial intelligence (AI) models such as ChatGPT to pick stocks? In their August 2023 paper entitled “ChatGPT-based Investment Portfolio Selection”, Oleksandr Romanko, Akhilesh Narayan and Roy Kwon explore use of ChatGPT to recommend 15, 30 or 45 S&P 500 stocks, with portfolio weights, based on textual sentiment as available to Chat GPT via web content up to September 2021. For robustness, they ask ChatGPT to repeat recommendations for each portfolios 30 times and select the 15, 30 or 45 most frequently recommended stocks for respective portfolios. They then test out-of-sample performance of the following five implementations of each portfolio during September 2021 to July 2023, mid-March 2023 to July 2023, and May 2023 to July 2023:

1. ChatGPT picks and ChatGPT weights.
2. ChatGPT picks weighted equally.
3. ChatGPT picks weighted based on minimum variance (Min Var) weights from a 5-year rolling weekly history.
4. ChatGPT picks weighted based on maximum return (Max Ret) weights from a 5-year rolling weekly history.
5. ChatGPT picks weighted based on maximum Sharpe ratio (Max Sharpe) weights from a 5-year rolling weekly history.

For benchmarking, they consider:

• Long-only portfolios that incorporate all possible combinations of 15, 30 or 45 S&P 500 stocks weighted as above for Min Var, Vax Ret or Max Sharpe.
• The S&P 500 Index, Dow Jones Industrial Average and the NASDAQ Index.
• Average performance of 13 popular equity funds.

Using weekly data as specified up to September 2021 for training and subsequent weekly data through June 2023 for out-of-sample testing, they find that:

Keep Reading

While prior research indicates that analyst forecasts of future stock returns are substantially biased upward, might the relative rankings of return forecasts be informative? In their June 2023 paper entitled “Analysts Are Good at Ranking Stocks”, Adam Farago, Erik Hjalmarsson and Ming Zeng apply within-analyst 12-month stock price targets to rank stocks in two ways:

1. Average Demeaned Return – each month, demean the returns implied by target prices from an analyst by subtracting from each return the average forecasted return for that analyst. Then, average the demeaned returns for a given stock across all analysts.
2. Average Ranking – each month, rank stocks by forecasted return for each analyst. Then, average the rankings for a given stock across all analysts covering that stock.

Both approaches remove the upward biases observed in raw target prices. To test analyst forecast informativeness, they then form hedge portfolios that are each month long (short) the equal-weighted or value-weighted fifths, or quintiles, of stocks with the highest (lowest) demeaned returns or rankings that month. Using 12-month target prices for each analyst who issues targets for at least three stocks during a month and associated monthly firm characteristics and stock prices during March 1999 through December 2021, they find that:

Keep Reading

How do returns of different asset classes recently interact with inflation as measured by monthly change in the not seasonally adjusted, all-items consumer price index (CPI) from the U.S. Bureau of Labor Statistics? To investigate, we look at lead-lag relationships between change in CPI and returns for each of the following 10 exchange-traded fund (ETF) asset class proxies:

• Equities:
  • SPDR S&P 500 (SPY)
  • iShares Russell 2000 Index (IWM)
  • iShares MSCI EAFE Index (EFA)
  • iShares MSCI Emerging Markets Index (EEM)
• Bonds:
  • iShares Barclays 20+ Year Treasury Bond (TLT)
  • iShares iBoxx $ Investment Grade Corporate Bond (LQD)
  • iShares JPMorgan Emerging Markets Bond Fund (EMB)
• Real assets:
  • Vanguard REIT ETF (VNQ)
  • SPDR Gold Shares (GLD)
  • Invesco DB Commodity Index Tracking (DBC)

Using monthly total CPI values and monthly dividend-adjusted prices for the 10 specified ETFs during December 2007 (limited by EMB) through June 2023, we find that: Keep Reading

During 1989 through 2019, the S&P 500 Index generated 5.5% real annual return, compared to just 2.5% annual real growth in U.S. gross domestic product (GDP). How can this disconnect happen? Can it continue? In the June 2023 version of his paper entitled “End of an Era: The Coming Long-Run Slowdown in Corporate Profit Growth and Stock Returns”, Michael Smolyansky examines interactions between U.S. stock market performance and declines in interest rates and corporate tax rates over the last three decades. He focuses on S&P 500 non-financial stocks adjusted for index additions/deletions and for changes in firm shares outstanding, allowing computation of per share metrics. He decomposes stock returns into: (1) change in price-earnings ratio (P/E);  (2) change in earnings before interest and taxes (EBIT); (3) change in interest expenses; and, (4) change in effective corporate tax rate. Using the specified annual data during 1962 through 2019, he finds that: Keep Reading

Which previously researched variables or combinations of such variables best predict long-term U.S. stock market returns? In their June 2023 paper entitled “Estimating Long-Term Expected Returns”, Rui Ma, Ben Marshall, Nhut Nguyen and Nuttawat Visaltanachoti compare abilities of several yield, yield/growth, valuation variables and combinations across these categories of variables to predict 10-year and 20-year U.S. stock market returns out-of-sample. Specifically, they test 25 predictors from the following individual variables and combinations thereof:

• Yield category: dividend yield, total yield, net total yield and cyclically adjusted total yield.
• Yield/growth category: current values of these yields plus historical earnings growth, dividend growth, total yield growth and cyclically adjusted total yield growth, respectively.
• Valuation category: total return cyclically adjusted price-earnings ratio, total wealth portfolio composition, equity market value-to-gross domestic product ratio (the Buffett indicator) and cyclical consumption.
• Combining categories based on: simple prediction average, inverse variance-weighted prediction average, constrained regression and Bayesian model averaging.

Their benchmark predictor is the historical average return. They use annualized log returns for all predictors, focusing on mean absolute errors and mean squared errors relative to actual future returns as accuracy metrics. They also consider also a mean-variance asset allocation perspective, allocating to the S&P 500 Index and 10-year U.S. Treasury notes to maximize gross Sharpe ratio based on predicted equity returns. Using monthly data as described above during 1871 through 2020, they find that:

Keep Reading

With monthly market state specified as positive (negative) when prior-month market excess return relative to U.S. Treasury bill (T-bill) yield is positive (negative), “Equity Factor Performance Following Positive and Negative Market Returns” reports that average monthly market excess return is 0.83% (10.0% annualized) positive market states and 0.05% (0.6% annualized) for negative states during August 1965 through January 2017. Is this finding reliable and easily exploitable? To check, we look at SPDR S&P 500 ETF Trust (SPY) monthly total returns after prior-month total returns are positive or negative out-of-sample with respect to the cited study. We also consider SPY excess returns according to whether its prior-month excess total returns are positive or negative. Using end-of-month SPY dividend-adjusted prices and monthly 3-month T-bill yield during January 2017 through June 2023, we find that:

Keep Reading

Do stock return anomalies perform differently after positive and negative monthly market returns? In their July 2023 paper entitled “The Market State, Mispricing and Asset Pricing Anomalies”, Michael Di Carlo and Ilias Tsiakas examine the role of the overall market state in estimating returns for stock return anomalies, specifying the market state as positive (negative) for a month when the market excess return relative to U.S. Treasury bill yield is positive (negative) the prior month. They then measure returns during each of the two states for 14 stock return anomalies, including: market beta, size, book-to-market, operating profitability, asset growth, momentum, short-term reversal, volatility, idiosyncratic volatility, correlation with the market, maximum return over the last month, maximum return over the past year, illiquidity and 1% value-at-risk. For each anomaly, they measure returns via a hedge portfolio that is each month long (short) the fifth, or quintile, of stocks with the highest (lowest) expected returns based on the relevant anomaly characteristic. Using the required monthly data for U.S. common stocks priced over $5 during August 1965 through January 2017, they find that: Keep Reading