Volatility Effects

Can long volatility investors improve performance of their portfolios by scaling positions inversely to some measure of volatility? In his March 2024 paper entitled “Volatility-Managed Volatility Trading”, Aoxiang Yang tests volatility risk premium (VRP) timing strategies that hold a volatility asset and a risk-free asset, with the weight of the former inverse to some measure of volatility. He considers three volatility assets that are each month:

1. Long a 1-month variance swap contract, held to maturity (with prices sometimes approximated using VIX-squared).
2. Long a 1-month constant-maturity VIX futures portfolio (ignoring both a margin requirement and frictions required to maintain constant maturity).
3. Short a 1-month constant-maturity S&P 500 Index at-the-money (ATM) straddle (including a margin requirement of 100% of selling proceeds plus 20% of current S&P 500 Index level, but ignoring frictions required to maintain constant maturity).

Each month, he weights each asset by one of four measures related to stock market volatility:

1. Inverse of realized volatility.
2. Inverse of implied volatility (VIX).
3. Inverse of an autoregression forecast of next-month volatility.
4. Forecast of next-month VRP (which has an inverse VIX term) from a vector autoregression of realized volatility and VIX.

For each measure of volatility, he multiplies by a scaling constant that makes the respective long volatility portfolio have the same standard deviation of monthly returns as the S&P 500 Index. His benchmark portfolios hold the same assets with constant weights. He further analyzes performance of volatility portfolios during times of high volatility (highest 20%) and low volatility (lowest 80%). Using estimates for actual monthly prices for variance swaps during 1990-2023 (and actual prices for recent subperiods), for a constant-maturity VIX futures portfolio during 2004-2023 and for a constant-maturity S&P 500 Index ATM straddles portfolio during 1996-2022, he finds that: Keep Reading

One concern about simple momentum strategies is data snooping bias impounded in selection of the lookback interval(s) used to measure asset momentum. To circumvent this concern, we consider the following argument:

• The CBOE Volatility Index (VIX) broadly indicates the level of financial markets distress and thereby the tendency of investors to act complacently (when VIX is low) or to act in panic (when VIX is high).
• Complacency translates to resistance in changing market outlook (long memory and lookback intervals), while panic translates to rapid changes of mind (short memory and short lookback intervals).
• The inverse of VIX is therefore indicative of the actual aggregate current lookback interval affecting investor actions.

We test this argument by:

• Setting a range for VIX using monthly historical closes from January 1990 through July 2002, before the sample period used for any tests of the Simple Asset Class ETF Momentum Strategy (SACEMS).
• Applying buffer factors to the bottom and top of this actual inverse VIX range to recognize that it could break above or below the historical range in the future.
• Segmenting the buffer-extended inverse VIX range into 12 equal increments and mapping these increments by rounding into momentum lookback intervals of 1 month (lowest segment) to 12 months (highest segment).
• Applying this same method to future end-of-month inverse VIX levels to select the SACEMS lookback interval for the next month.

We test the top one (Top 1), the equally weighted top two (EW Top 2) and the equally weighted top three (EW Top 3) SACEMS portfolios. We focus on compound annual growth rate (CAGR), maximum drawdown based on monthly measurements, annual returns and Sharpe ratio as key performance statistics. To calculate excess annual returns for the Sharpe ratio, we use average monthly yield on 3-month Treasury bills during a year as the risk-free rate for that year. Benchmarks are these same statistics for tracked SACEMS. Using monthly levels of VIX since inception in January 1990 and monthly dividend-adjusted prices of SACEMS assets since February 2006 (initial availability of a commodities ETF), all through January 2024, we find that: Keep Reading

In response to “Shorting VXX with Crash Protection”, which investigates shorting iPath S&P 500 VIX Short-Term Futures (VXX) to capture the equity volatility risk premium, a subscriber asked about instead using a long position in ProShares Short VIX Short-Term Futures (SVXY). To investigate, we consider two scenarios based on monthly measurements:

1. Buy and Hold – buy an initial amount of SVXY and let this position ride indefinitely. This is a long-term investment strategy.
2. Monthly Skim – buy the same initial amount of SVXY and move to SPDR Bloomberg 1-3 Month T-Bill ETF (BIL) any month-end gains over the initial investment (the beginning-of-month SVXY position may become smaller, but not larger, than the initial investment). This is more an income-generating investment strategy.

The offeror changed the SVXY investment objective at the end of February 2018 (when short VIX strategies crashed), more conservatively targeting henceforth -0.5 times the daily performance of the S&P 500 VIX Short-Term Futures Index rather than -1.0 times as before. We therefore examine SVXY performance separately before and after that change. We assume switching frictions of 0.25% for movements of funds from SVXY to BIL in scenario 2. Using monthly adjusted closing prices for SVXY and BIL during October 2011 through December 2023, we find that: Keep Reading

Academic studies of stock portfolio optimization often use an equal-weighted (EW) strategy as benchmark. Are there simple EW enhancements that researchers ought to consider instead? In their December 2023 paper entitled “Outperforming Equal Weighting”, Antonello Cirulli and Patrick Walker test three sets of enhanced long-only EW portfolios relying solely on past returns:

1. Momentum-enhanced EW – sort stocks into tenths (deciles) from lowest to highest average weekly return over the last 12 months.
2. Volatility-enhanced EW – sort stocks into deciles from highest t0 lowest standard deviation of weekly returns over the last five years.
3. Sharpe ratio-enhanced EW – sort stocks into deciles from lowest to highest Sharpe ratio calculated with weekly returns over the last years.

For each set, they then exclude the bottom 1, 2, 3, 4 or 5 deciles and weight stocks in retained deciles equally for a total of 15 enhanced EW portfolios. They reform all portfolios on the first Wednesday of each month. They then compare net performances of these portfolios to those of simple EW and capitalization-weighted portfolios of all stocks in the universe after debiting 0.1% frictions for turnover. They focus on large-capitalization/liquid stocks and check robustness of findings to subperiods, lookback intervals, level of frictions and rebalancing frequency. Using weekly returns in U.S. dollars, adjusted for splits and dividends, of MSCI USA, Europe, Emerging Markets and Developed Markets stocks starting five years before the test period of April 2002 through March 2022, they find that:

Keep Reading

Does market volatility predictably affect returns to simple moving average (SMA) trend-following strategies? In their November 2023 paper entitled “Market Volatility and the Trend Factor”, Ming Gu, Minxing Sun, Zhitao Xiong and Weike Xu investigate how stock market volatility affects multi-SMA trend factor profitability. They first assess significance of the trend factor premium, as follows:

• For each stock at the close on the last trading day of each month:
  • Compute SMAs of prices for lookback intervals of 3, 5, 10, 20, 50, 100, 200, 400, 600, 800 and 1000 trading days, and divide each SMA by the end price.
  • Starting five years into the sample period (1931), regress next-month stock returns on corresponding monthly SMA ratios over the past 60 months.
  • Average the SMA ratio regression coefficients separately over the past 12 months to estimate next-month coefficients and apply these coefficients to estimate next-month return.
• At the end of each month, sort all stocks into tenths, or deciles, based on estimated next-month returns and form a trend factor hedge portfolio that is long (short) the equal-weighted top (bottom) decile. The trend factor premium is the monthly gross return for this portfolio.

They then assess how trend factor hedge portfolio returns interact with monthly stock market return volatility (standard deviation of monthly value-weighted market returns over the past 12 months) by specifying volatility has high or low when its prior-month value is above or below the full-sample median. Using data for all listed U.S. common stocks, excluding those priced below $5 or in the lowest tenth of NYSE market capitalizations, during January 1926 through December 2022, they find that:

Keep Reading

Do Relative Rotation Graphs (RRG), which visually segregate assets into leading, weakening, lagging or improving quadrants by relative performance, effectively identify equity sectors with relatively strong future returns? In his September 2023 paper entitled “Dynamic Sector Rotation”, John Rothe tests an RRG-based sector relative momentum strategy with stop-loss risk management based on volatility. Specifically, he:

• Selects a universe of 31 sector sector/subsector exchange-traded funds (ETFs) based on daily trading volume, years in existence, overlap with other sector/subsectors, assets under management and liquidity.
• Each week, holds the equal-weighted top 5 ETFs crossing into the RRG improving quadrant.
• Manages the risk of each holding continuously via a Wilder Volatility Stop with a 5-day range.
• Assumes a 2% annual management fee.

His benchmark is the S&P 500 Momentum Index. Using weekly returns for the selected ETF universe during a test period spanning January 2013 through mid-2023, he finds that: Keep Reading

Is dispersion of stock return forecasts from different machine learning models trained on the same history (as a proxy for variation in human beliefs) a useful predictor of stock returns? In their August 2023 paper entitled “Machine Forecast Disagreement”, Turan Bali, Bryan Kelly, Mathis Moerke and Jamil Rahman relate dispersion in 100 monthly stock return predictions for each stock generated by randomly varied versions of a machine learning model applied to 130 firm/stock characteristics. They measure machine return forecast dispersion for each stock as the standard deviation of predicted returns. They then each month sort stocks into tenths (deciles) based on this dispersion, form either a value-weighted or an equal-weighted portfolio for each decile and compute average next-month portfolio return. Their key metric is average next-month return for a hedge portfolio that is each month long (short) the stocks in the lowest (highest) decile of machine return forecast dispersions. Using the 130 monthly firm/stock characteristics and associated monthly stock returns for a broad sample of U.S. common stocks (excluding financial and utilities firms and stocks trading below $5) during July 1966 through December 2022, they find that:

Keep Reading

How have different asset classes recently interacted with the CBOE Volatility Index (VIX)? To investigate, we look at lead-lag relationships between VIX 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)

We look also at average next-month performances of these ETFs across ranges of of a VIX 3-month simple moving average (SMA3). Using end-of-month levels of VIX since January 1990 and dividend-adjusted monthly closing prices for the asset class proxies as available since July 2002, all through July 2023, we find that: Keep Reading

A subscriber requested comparison of four variations of an “Ivy 5” asset class allocation strategy, as follows:

1. Ivy 5 EW: Assign equal weight (EW), meaning 20%, to each of the five positions and rebalance annually.
2. Ivy 5 EW + SMA10: Same as Ivy 5 EW, but take to cash any position for which the asset is below its 10-month simple moving average (SMA10).
3. Ivy 5 Volatility Cap: Allocate to each position a percentage up to 20% such that the position has an expected annualized volatility of no more than 10% based on daily volatility over the past month, recalculated monthly. If under 20%, allocate the balance of the position to cash.
4. Ivy 5 Volatility Cap + SMA10: Same as Ivy 5 Volatility Cap, but take completely to cash any position for which the asset is below its SMA10.

To perform the tests, we employ the following five asset class proxies:

iShares 7-10 Year Treasury Bond ETF (IEF)
SPDR S&P 500 ETF Trust (SPY)
Vanguard Real Estate Index Fund (VNQ)
iShares MSCI EAFE ETF (EFA)
Invesco DB Commodity Index Tracking Fund (DBC)

We consider monthly performance statistics, annual performance statistics, and full-sample compound annual growth rate (CAGR) and maximum drawdown (MaxDD). Annual Sharpe ratio uses average monthly yield on 3-month U.S. Treasury bills (T-bills) as the risk-free rate. The DBC series in combination with the SMA10 rule are limiting with respect to sample start date and the first return calculations. Using daily and monthly dividend-adjusted closing prices for the five asset class proxies and T-bill yield as return on cash during February 2006 through July 2023, we find that:

Keep Reading

Is the ability of the VIX percentile threshold rule described in “Using VIX and Investor Sentiment to Explain Stock Market Returns” to explain future stock market excess return in-sample readily exploitable out-of-sample? To investigate, we test a strategy (VIX Percentile Strategy) that each month holds SPDR S&P 500 ETF Trust (SPY) or 3-month U.S. Treasury bills (T-bills) according to whether a recent end-of-month level of the CBOE Volatility Index (VIX) is above or below a specified inception-to-date (not full sample) percentage threshold. To test sensitivities of the strategy to settings for its two main features, we consider:

• Each of 70th, 75th, 80th, 85th or 90th percentiles as the VIX threshold for switching between T-bills and SPY.
• Each of 0, 1, 2 or 3 skip months between VIX measurement and strategy response.

We focus on compound annual growth rate (CAGR) and maximum drawdown (MaxDD) as essential performance metrics and use buy-and-hold SPY as a benchmark. We do not quantify frictions due to switching between SPY and T-bills for the VIX Percentile Strategy. Using end-of-month VIX levels since January 1990 and dividend-adjusted SPY prices and T-bill yields since January 1993 (SPY inception), all through May 2023, we find that: Keep Reading