Equity Premium

Does finance professor David Kass, who presents annual lists of stock picks on Seeking Alpha, make good selections? To investigate, we consider his picks of “10 Stocks for 2020”, “16 Stocks For 2021”, “12 Stocks For 2022”, “10 Stocks For 2023” and “10 Stocks For 2024”. For each year and each stock, we compute total (dividend-adjusted) return. For each year, we then compare the average (equal-weighted) total return for a David Kass portfolio to that of  SPDR S&P 500 ETF Trust (SPY). Using dividend-adjusted returns from Yahoo!Finance for SPY and most stock picks and returns from Barchart.com and Investing.com for three picks during their selection years, we find that: Keep Reading

Do stock-by-stock return forecasts from deep learning produce an exploitable aggregate equity risk premium (ERP) forecast? In the January 2025 revision of their paper entitled “The Aggregated Equity Risk Premium”, Vitor Azevedo, Christoph Riedersberger and Mihail Velikov predict ERP by first applying deep learning to predict returns for individual U.S. stocks and then aggregating these returns at the market level. The firm-level forecasts come from combined outputs of several neural networks of varying complexity applied to 290 firm-level characteristics, 14 U.S. economic variables and 49 industry classification indicators. They iterate these forecasts annually using an expanding training window and a rolling six-year validation window. For comparison, they consider some conventional ERP forecasting approaches. They quantify the economic value of aggregate ERP forecasts via a stock market timing strategy that each month allocates to stocks or U.S. Treasury bills with a 50% leverage limit and conservative 0.5% portfolio rebalancing frictions. Using the specified inputs during March 1957 through December 2021 (with out-of-sample testing commencing January 2000), they find that:

Keep Reading

How do exchange-traded-funds (ETF) focused on development of robotics-artificial intelligence (AI), an arguably hot area of technology, perform? To investigate, we consider eight of the largest such ETFs, all currently available, as follows:

• • Robo Global Robotics and Automation Index ETF (ROBO)
  • ARK Autonomous Technology & Robotics ETF (ARKQ)
  • Global X Robotics & Artificial Intelligence ETF (BOTZ)
  • First Trust Nasdaq Artificial Intelligence and Robotics ETF (ROBT)
  • Global X Artificial Intelligence & Technology ETF (AIQ)
  • iShares Future AI & Tech ETF (ARTY)
  • Spear Alpha ETF (SPRX)
  • WisdomTree Artificial Intelligence and Innovation Fund (WTAI)

We use Invesco QQQ Trust (QQQ) as a benchmark, assuming investors look at robotics-AI stocks as a way to beat other technology stocks. We focus on monthly return statistics, along with compound annual growth rates (CAGR) and maximum drawdowns (MaxDD). Using monthly total returns for the eight robotics-AI ETFs and QQQ as available through December 2024, we find that: Keep Reading

Past research relating U.S. stock market returns to the party holding the Presidency mostly concludes that Democratic presidents are better for the stock market than Republican presidents. However, Presidents share power conferred by the electorate with Congress. Does historical data confirm that Democratic control of Congress is also better for stock market returns than Republican control of Congress? Is control of the smaller Senate more decisive than control of the House of Representatives? To check, we relate annual U.S. stock market (S&P 500 Index) returns to various combinations of party control of the Presidency, the Senate and the House of Representatives. Using party in power data and annual levels of the S&P 500 Index for December 1927 through December 2024 (97 years), we find that: Keep Reading

How should investors think about research using long-run financial data? In their October 2024 paper entitled “Long-Run Asset Returns”, David Chambers, Elroy Dimson, Antti Ilmanen and Paul Rintamäki survey the body of evidence on historical return premiums for stocks, bonds, real estate and commodities over the current and previous two centuries. They discuss benefits and pitfalls of long-run datasets and make suggestions on best practices. They also compare premium estimates from alternative data compilers. Based on the body of long-run asset class return research, they conclude that:

Keep Reading

Are the human choices in studies that apply machine learning models to forecast stock returns critical to findings? In other words, is there a confounding machine learning design choices zoo? In their November 2024 paper entitled “Design Choices, Machine Learning, and the Cross-section of Stock Returns”, Minghui Chen, Matthias Hanauer and Tobias Kalsbach analyze effects of varying seven key machine learning design choices: (1) machine learning model used, (2) target variable/evaluation metric, (3) target variable transformation (continuous or discrete dummy), (4) whether to use anomaly inputs from pre-publication subperiods or not, (5) whether to compress correlated features, (6) whether to sue a rolling or expanding training window and (7) whether to include micro stocks in the training sample. They examine all possible combinations of these choices, resulting in 1,056 machine learning models. For each machine learning model each month, they:

1. Rank stocks on each of 207 potential return predictors and map rankings into [-1, 1] intervals. In case of missing inputs, they set the ranking value to 0.
2. Apply rankings to predict a next-month target variable (return in excess of the risk-free rate, market-adjusted return or 1-factor model risk-adjusted return) for each stock with market capitalization above a 20% NYSE threshold during January 1987 through December 2021.
3. Reform a hedge portfolio that is long (short) the value-weighted tenth, or decile, of stocks with the highest (lowest) predicted target variable and compute next-month portfolio return.

Using monthly data as available for all listed U.S. common stocks during January 1957 through December 2021, they find that: Keep Reading

Can investors safely use leverage to squeeze incremental return from low-volatility/factor-tilted stocks, thereby avoiding underperformance of these stocks during bull markets? In their October 2024 paper entitled “Low-Risk Alpha Without Low Beta”, David Blitz, Clint Howard, Danny Huang and Maarten Jansen exploit the low-volatility anomaly by leveraging multifactor, low-risk, global stock portfolios to a beta of 1.0 while controlling tracking error relative to a capitalization-weighted benchmark. Their portfolio formation rules are:

• The portfolio is long only and fully invested in liquid (large-capitalization) stocks.
• Maximum individual stock weight is the lower of 1.5% or 20 times its benchmark weight.
• Exposure to countries, regions and sectors may deviate at most 10% from benchmark weights.
• Portfolio beta (portfolio-weighted sum of historical stock betas for the last 156 weekly returns) must be less than 0.8 relative to the benchmark.
• Portfolio optimization involves trading off expected returns, benchmark tracking error and turnover. Expected stock returns derive from a multifactor score with 50% for low-risk (equal-weighted combination of past 260-day volatility, 156-week volatility, 260-day beta and 156-week beta), 16.67% for value (net payout yield), 16.67% for quality (gross profits to assets) and 16.67% for momentum (return from 12 months ago to one month ago).
• Use synthetic positions (for example, via equity options) to achieve leverage, with no cash collateral and financing costs equal to the risk-free rate.
• Rebalance at the end of each month but ignore slight deviations from target weights.

They separately discuss impacts of portfolio rebalancing frictions and additional leverage costs/penalties. They focus on developed markets but also look at an emerging markets sample and North American, European and Asia Pacific subsamples. Using daily and monthly data for developed market stocks since December 1985 and emerging market stocks since December 1995, all through December 2023, along with contemporaneous spreads and interest/Treasury bill rates, they find that: Keep Reading

Can investors reasonably anticipate the signals (stock rankings) for stock anomalies that are based on firm financial information. In their August 2024 paper entitled “Predicting Anomalies”, Boone Bowles, Adam Reed, Matthew Ringgenberg and Jake Thornock investigate whether: (1) stock returns follow predictable patterns before availability of anomaly trading signals; and, (2) anomaly trading signals are themselves predictable. They focus on a set 28 published anomalies that are entirely based on publicly available information in quarterly financial statements. They each quarter for each anomaly reform a hedge portfolio that is long (short) the tenth of stocks with the highest (lowest) expected returns. They consider four models to predict stock rankings for each anomaly: (1) a first-order autoregression that projects strength of signals; (2) a first-order autoregression that projects stock rankings; (3) a machine learning model that uses past anomaly signals and rankings; and, (4) a (martingale) model that assumes anomaly portfolio rankings for next quarter will be the same as current rankings. Using as-published specifications for each of the 28 anomalies plus daily returns and quarterly/annual financial reports for a broad sample of U.S. stocks during January 1990 through December 2019, they find that: Keep Reading

Can market efficiency be falling despite ubiquitous data, computing and networking? In his August 2024 paper entitled “The Less-Efficient Market Hypothesis”, Clifford Asness argues that markets have become less efficient in the relative pricing of common stocks over recent decades. To make his argument, he relies on the ratio of expensive stock valuations to cheap stock valuations (the value spread). He considers two versions of this spread, one based on the conventional price-to-book ratio to measure value and the other based on five industry-neutral value metrics. He discusses three potential reasons why the value spread is rising. He closes with advice for value investors. Reflecting on 35 years of research experience, he concludes that:

Keep Reading

Why have stock return anomalies generally degraded over recent decades? In their August 2024 paper entitled “What Drives Anomaly Decay?”, Jonathan Brogaard, Huong Nguyen, Tālis Putniņš and Yuchen Zhang examine why stock return anomalies decay by:

• Decomposing returns into market-wide, public firm-specific and private firm-specific elements.
• Separating cash flow and discount rate effects within each of these three components.
• Accounting for noise.

This breakdown lets them determine whether changes in anomaly returns over time derive from anomaly publication, identifiable liquidity shocks (such as stock price decimalization) or a more general increase market efficiency. They apply this approach to daily returns of long-short (hedge) portfolios, reformed monthly, for 204 stock return anomalies from Open Source Asset Pricing. Using the required firm characteristics and daily prices for all NYSE/AMEX/NASDAQ common stocks during 1956 through 2021 (an average 4,029 firms per year and a total of 16,966 firms), they find that:

Keep Reading