Equity Premium

Can machine learning exploit interactions between many equity factors and many potential factor return predictors to create an attractive factor timing strategy? In their August 2024 paper entitled “Optimal Factor Timing in a High-Dimensional Setting”, Robert Lehnherr, Manan Mehta and Stefan Nagel apply machine learning with mean-variance optimization to time equity factors when the numbers of factors and potential factor return predictors are large. They consider both a small set of four (size, book-to-market, profitability and investment) and a much larger set of 131 factors. They focus on a small set of 11 potential predictors (five economic variables and six specific to the small set of factors) but consider also a much larger set augmenting those 11 with many other factor specific variables. They simplify outputs by suppressing the weakest signals. Their machine learning process each year uses an expanding window of at least 20 years for training, one year for validation and one year for testing. They focus on Sharpe ratio as the essential performance metric. Their benchmarks are annually rebalanced: (1) equal weighting of factors; and, (2) straightforward mean-variance optimization of factors that ignores interactions between factors and potential predictors. To estimate net performance, they apply 0.1% portfolio reformation frictions at the portfolio of factors (not factor) level. Using monthly factor portfolio returns and economic variable values during January 1965 through December 2022, with 1986 the first year of portfolio testing, they find that: Keep Reading

How has the immediate price impact associated with a stock entering or leaving the S&P 500 evolved? In the March 2024 revision of their paper entitled “The Disappearing Index Effect”, Robin Greenwood and Marco Sammon revisit abnormal returns during the trading day after S&P 500 additions and deletions and investigate four potential drivers of findings. Using announcement dates, implementation dates and daily returns for 732 S&P 500 additions and 726 S&P 500 deletions, and holdings of large U.S. equity funds, during 1980 through 2020, they find that:

Keep Reading

Do stock markets following predictable long boom and bust periods? In the August 2024 draft of their paper entitled “The Anatomy of Lost Stock Market Decades”, Todd Feldman and Brian Yang examine the regularity/frequency of bull periods (strong gains) and lost periods (no gains) of at least 10 years. They also test two metrics to identify when the S&P 500 Index is in a bull or lost period: (1) the ratio of the S&P 500 Index level to a dividend discount model (DDM) valuation of the index; and, (2) an exponential cumulative loss metric calibrated via a 20-year moving average (weighting recent losses more than older losses to sharpen regime shift detection). Using monthly stock market levels from Global Financial Data for the U.S., Canada, Japan, Australia, Germany and France and Robert Schiller’s data for the S&P Composite Index from the 1800s through 2023, they find that:

Keep Reading

Are preferred stock strategies, as implemented by exchange-traded funds (ETF), attractive? To investigate, we consider seven of the largest preferred stock ETFs, all currently available, in order of longest to shortest available histories:

• Invesco Financial Preferred (PGF) – generally invests at least 90% of assets in preferred securities of financial institutions.
• iShares Preferred and Income Securities (PFF) – tracks an index composed of U.S. dollar-denominated preferred and hybrid securities.
• Invesco Preferred (PGX) – tracks performance of fixed-rate preferred securities issued in the U.S.
• SPDR ICE Preferred Securities (PSK) – seeks exposure to non-convertible preferred securities with par $25 and aggregate minimum par value $250 million.
• VanEck Vectors Preferred Securities ex Financials (PFXF) – tracks performance of U.S. exchange-listed hybrid debt, preferred stock and convertible preferred stock issued by non-financial corporations.
• First Trust Preferred Securities and Income (FPE) – normally invests at least 80% of assets (including borrowings) in preferred securities and income-producing debt securities, including corporate bonds, high-yield securities and convertible securities.
• Global X U.S. Preferred (PFFD) – invests in a broad basket of U.S. preferred stocks.

We use a monthly rebalanced portfolio of 60% SPDR S&P 500 (SPY) and 40% iShares iBoxx $ Investment Grade Corporate Bond (LQD) (60-40) as a simple hybrid benchmark for all these funds except PGF, for which we use Financial Select Sector SPDR (XLF). We focus on monthly return statistics, along with compound annual growth rates (CAGR) and maximum drawdowns (MaxDD). Using monthly returns for the preferred stock ETFs and benchmarks as available through August 2024, we find that: Keep Reading

Are equity multifactor strategies, as implemented by exchange-traded funds (ETF), attractive? To investigate, we consider eight multifactor ETFs, all currently available:

• iShares Edge MSCI Multifactor USA (LRGF) – holds large and mid-cap U.S. stocks with focus on quality, value, size and momentum, while maintaining a level of risk similar to that of the market. The benchmark is iShares Russell 1000 (IWB).
• iShares Edge MSCI Multifactor International (INTF) – holds global developed market ex U.S. large and mid-cap stocks based on quality, value, size and momentum, while maintaining a level of risk similar to that of the market. The benchmark is iShares MSCI ACWI ex US (ACWX).
• Goldman Sachs ActiveBeta U.S. Large Cap Equity (GSLC) – holds large U.S. stocks based on good value, strong momentum, high quality and low volatility. The benchmark is SPDR S&P 500 (SPY).
• John Hancock Multifactor Large Cap (JHML) – holds large U.S. stocks based on smaller capitalization, lower relative price and higher profitability, which academic research links to higher expected returns. The benchmark is SPY.
• John Hancock Multifactor Mid Cap (JHMM) – holds mid-cap U.S. stocks based on smaller capitalization, lower relative price and higher profitability, which academic research links to higher expected returns. The benchmark is SPDR S&P MidCap 400 (MDY).
• JPMorgan Diversified Return U.S. Equity (JPUS) – holds U.S. stocks based on value, quality and momentum via a risk-weighting process that lowers exposure to historically volatile sectors and stocks. The benchmark is SPY.
• Xtrackers Russell 1000 Comprehensive Factor (DEUS) – seeks to track, before fees and expenses, the Russell 1000 Comprehensive Factor Index, which seeks exposure to quality, value, momentum, low volatility and size factors. The benchmark is IWB.
• Vanguard U.S. Multifactor (VFMF) – uses a rules-based quantitative model to evaluate U.S. common stocks and construct a U.S. equity portfolio that seeks to achieve exposure to multiple factors across market capitalizations (large, mid and small). The benchmark is iShares Russell 3000 (IWV).

We focus on monthly return statistics, along with compound annual growth rates (CAGR) and maximum drawdowns (MaxDD). Using monthly returns for the seven equity multifactor ETFs and benchmarks as available through August 2024, we find that: Keep Reading

How does the performance of the global multi-class market look when evaluated at a monthly frequency? In their August 2024 paper entitled “The Risk and Reward of Investing”, Ronald Doeswijk and Laurens Swinkels assess global investing rewards and risks via an exhaustive $150 trillion portfolio of investable global assets priced at a monthly frequency, enabling greater granularity of risk estimates than does the annual frequency used in prior research. They consider five asset classes: equities, real estate, non-government bonds, government bonds and commodities. For these classes and the multi-class market, they examine stability of Sharpe ratios and severity, frequency and duration of drawdowns. Their default base currency is the U.S. dollar, but they measure effects of choosing one of nine other currencies on global market portfolio performance. They calculate excess investment returns generally relative to government bill yields as a proxy for return on savings. Using monthly returns for all investable global assets with reinvested dividends during 1970 through 2022, they find that:

Keep Reading

Do exchange-traded funds (ETF) focused on cybersecurity stocks offer attractive performance? To investigate, we compare performance statistics of six cybersecurity ETFs, all currently available, to those of Invesco QQQ Trust (QQQ), as follows:

1. Amplify Cybersecurity ETF (HACK)
2. First Trust NASDAQ Cybersecurity ETF (CIBR)
3. iShares Cybersecurity and Tech ETF (IHAK)
4. Global X Cybersecurity ETF (BUG)
5. ProShares Ultra Nasdaq Cybersecurity (UCYB)
6. WisdomTree Cybersecurity Fund (WCBR)

We focus on average return, standard deviation of returns, reward/risk (average return divided by standard deviation of returns), compound annual growth rate (CAGR) and maximum drawdown (MaxDD), all based on monthly data. Using monthly dividend-adjusted returns for all specified ETFs since inceptions and for QQQ over matched sample periods, all through July 2024, we find that: Keep Reading

“Nixed: The Upside of Getting Dumped”, flagged by a subscriber, finds that “index deletions…could add an abnormal upside to a portfolio when the current growth-dominated bubble starts to deflate.” The authors have quantified findings as the Research Affiliates Deletions Index (NIXT), constructed by:

1. Starting with deletions due to market capitalization changes from the 500 and 1,000 largest U.S. stocks by market capitalization.
2. Removing the bottom 20% of deletions based on firm quality assessments.
3. Holding the equal-weighted remaining deletions up to five years (or until they rejoin a top market capitalization index), rebalancing annually at the end of May.

Do index deletions inherently underperform? To investigate we look at stocks deleted from the S&P 500 Index due to market capitalization changes over the past few years and compare their performances to that of SPDR S&P 500 ETF Trust (SPY) since the close on respective deletion dates. Using dividend-adjusted prices for 43 S&P 500 deletions at closes on deletion dates and corresponding dividend-adjusted prices for SPY since during April 2020 through 8/23/24, we find that: Keep Reading

A subscriber requested measurement of a “premium” associated with stocks of innovative technology firms by looking at the difference in returns between the following two exchange-traded funds (ETF):

• Invesco QQQ Trust (QQQ), which generally tracks the NASDAQ 100 Index
• SPDR S&P 500 ETF Trust (SPY)

Using monthly dividend-adjusted closing prices for these ETFs during March 1999 (limited by QQQ) through July 2024, we find that: Keep Reading

Does the combination of an artificial intelligence (AI)-generated financial news sentiment with a complex financial stress metric generate good stock market timing signals? In their April 2024 paper entitled “Mixing Financial Stress with GPT-4 News Sentiment Analysis for Optimal Risk-On/Risk-Off Decisions”, Baptiste Lefort, Eric Benhamou, Jean-Jacques Ohana, David Saltiel, Beatrice Guez and Thomas Jacquot devise and test a risk-on/risk-off strategy for stock market timing. The strategy combines:

• Stress Index (SI): based on VIX, TED spread, Credit Default Swap index and realized volatilities across major equity, bond and commodity markets, all normalized and then aggregated by category. Overall SI is the average of category results, rescaled to fall between 0 and 1.
• News sentiment signal: 10-day moving average of ChatGPT 4 assessments of the sentiment (1 for positive or 0 for negative) in Bloomberg daily market summaries.

They consider six strategies:

1. Benchmark (or Long Only) – buy and hold the index, with constant volatility scaling to match the final (retrospective) volatility of an active strategy.
2. VIX – weight the stock index according to VIX, with times of stress indicated by VIX above its 80th percentile.
3. SI – weight the stock index according to the value of SI as described above.
4. News – weight the stock index according to the ChatGPT 4 news sentiment signal.
5. SI News – weight the stock index according to the product of SI and News.
6. Dynamic SI News – because SI News either significantly outperforms or underperforms SI alone during subperiods, each month weight the stock index according to either SI alone or SI News, whichever has the better Sharpe ratio over the past 250 trading days at the end of the prior month.

They test the strategy on the S&P 500 Index, the NASDAQ 100 Index and an equal-weighted combination of these two indexes plus the Nikkei 225, Euro Stoxx 50 and Emerging Markets indexes. They assume trading frictions of 0.2% of value traded. Using daily values of all specified inputs during January 2005 through December 2023, they find that: Keep Reading