Technical Trading

Does technical trading work, or not? Rationalists dismiss it; behavioralists investigate it. Is there any verdict? These blog entries relate to technical trading.

Page 1 of 2212345678910...Last »

Long-run Moving Average Horse Race for Timing the U.S. Stock Market

Does timing the U.S. stock market with moving averages work? In his October 2015 paper entitled “A Comprehensive Look at the Real-Life Performance of Moving Average Trading Strategies”, Valeriy Zakamulin employs a very long dataset to estimate out-of-sample performance and robustness (subsample performance) of four distinct technical trading rules. Specifically, he seeks answers to the following questions:

  • How well does market timing really work?
  • Does overweighting or underweighting recent prices improve market timing?
  • Do timing rules have optimal lookback intervals?
  • Can timing rules accurately exploit bull and bear market states?

The four trading rules are:

  1. Momentum (MOM) – final price minus initial price across the measurement interval.
  2. Price minus Simple Moving-Average (P-SMA) – final price minus linearly decreasing weighted average of past prices backward over the measurement interval.
  3. Price minus Reverse Exponential Moving Average (P-REMA) – final price minus exponentially decreasing weighted average of past prices with decay factor 0.8, for an effect between MOM and P-SMA.
  4. Double-Crossover Method (DCM) – long-interval EMA minus short-interval EMA with decay factors 0.8 and the short interval fixed at two months.

For all four rules, a positive (negative or zero) signal means hold stocks (the risk-free asset) the following month. For optimization of moving average lookback intervals, he considers both rolling 10-year windows and inception-to-date (expanding window) data and tests intervals up to 24 months. His total sample spans 1860 through 2014, with the first 10 years reserved for lookback interval optimization. He also considers two equal subsamples (1860-1942 and 1932-2014), with the first 10 years of each reserved for initial optimization. He assumes one-way switching friction 0.25%. He uses several risk-adjusted performance measures, emphasizing Sharpe ratio. Using monthly capital gains and total returns of the S&P Composite stock price index and the contemporaneous U.S. Treasury bill yield as the risk-free rate during January 1860 through December 2014, he finds that: Keep Reading

Exploiting the Trend Lag of Small Stocks?

Do small capitalization stocks exploitably lag broad market trends? In their October 2015 paper entitled “Slow Trading and Stock Return Predictability”, Matthijs Lof and Matti Suominen investigate whether overall stock market trends predict variation in the size effect and therefore the performance of small capitalization exchange-traded funds (ETF). For size effect testing, they each year at the end of June rank stocks into tenths (deciles) by market capitalization and calculate the size effect as the difference in value-weighted average returns between the smallest and largest deciles. Using daily returns, trading volumes and institutional buying and selling data for a broad sample of U.S. common stocks during 1964 through 2014 and for a selection of small capitalization ETFs as available through 2014, they find that: Keep Reading

Valuation/Trend Hedging of a Value and Momentum Stock Portfolio

Is there a way to suppress the volatility and drawdowns of a mixed value and momentum stock strategy while retaining most of its benefit? In his September 2015 paper entitled “Learning to Play Offense and Defense: Combining Value and Momentum from the Bottom up, and the Top Down”, Mebane Faber examines the feasibility of a strategy that combines market valuation and market trend timing (defense) with a mixed value and momentum stock selection strategy (offense). Specifically:

For offense, he each month: (1) ranks stocks by each of price-to-earnings, price-to-book and earnings before interest and taxes-to-total enterprise value ratios and then re-ranks them by the average of the three separate value rankings; (2) ranks stocks by each of 3-month, 6-month and 12-month past returns and then re-ranks them by the average of the three separate momentum rankings; and, (3) forms an equally weighted portfolio of the top 100 value and top 100 momentum stocks and holds for three months (three overlapping portfolios).

For defense, he each month: (1) hedges half of the portfolio by shorting the S&P 500 Index if the long-term real earnings yield for the S&P 500 (inverse of the Cyclically Adjusted Price-Earnings ratio, CAPE or P/E10 as calculated by Robert Shiller, minus the most recently available actual 12-month U.S. inflation rate) is in the 20% of its lowest inception-to-date monthly values; and, (2) hedges half of the portfolio by shorting the S&P 500 Index if the index is below its 12-month simple moving average. 

The overall portfolio can therefore be 100% long “offense” stocks, 50% hedged or market neutral. He does not account for costs of portfolio reformations or hedging. Using monthly total returns for all NYSE stocks in the top 60% of market capitalizations, monthly levels of the S&P 500 Total Return Index and monthly values of CAPE during 1964 through 2014, he finds that: Keep Reading

Exploiting Stock Limit Order Books?

Do stock limit order books tip the direction of stock price? In their October 2015 paper entitled “Enhancing Trading Strategies with Order Book Signals”, Alvaro Cartea, Ryan Donnelly and Sebastian Jaimungal test the use of buying and selling pressures based on limit order book data to predict the direction, depth and magnitude of near-term stock price movements. They define pressure simply as the difference between the volume of limit orders at the highest bid minus volume of limit orders at the lowest ask, divided by the sum of the two volumes. When the ratio approaches 1 (-1), there is strong buying (selling) pressure. They test the degree to which traders can enhance performance of round-trip trading strategies by exploiting buying and selling pressure. Using stock limit order book data for ten Nasdaq stocks with relatively large tick sizes during January 2014 through June 2014 to calibrate trading rules and during July 2014 through December 2014 to test application of the rules to trading, they find that: Keep Reading

Simple Tests of Sy Harding’s Seasonal Timing Strategy

Several readers have inquired over the years about the performance of Sy Harding’s Street Smart Report Online (now unavailable due to Mr. Harding’s death), which included the Seasonal Timing Strategy. This strategy combines “the market’s best average calendar entry [October 16] and exit [April 20] days with a technical indicator, the Moving Average Convergence Divergence (MACD).” According to Street Smart Report Online, applying this strategy to a Dow Jones Industrial Average (DJIA) index fund generated a cumulative return of 213% during 1999 through 2012, compared to 93% for the DJIA itself. For robustness testing, we apply this strategy to SPDR S&P 500 (SPY) since its inception and consider several alternatives, as follows:

  1. SPY – buy and hold SPY.
  2. Seasonal-MACD – seasonal timing with MACD refinement.
  3. Seasonal Only – seasonal timing without MACD refinement.
  4. SMA200 – hold SPY (13-week U.S. Treasury bills (T-bills) when the S&P 500 Index is above (below) its 200-day simple moving average at the prior daily close. 

Using daily closes for the S&P 500 Index, daily dividend-adjusted closes for SPY and daily T-bill yields during 1/29/93 (SPY inception) through 9/25/15, we find that: Keep Reading

Annual Stock Market Streaks

Are annual stock market winning and losing streaks informative about future market performance? To investigate, we consider up and down annual streaks for the Dow Jones Industrial Average (DJIA). We look at streaks in two ways:

  1. Retrospective (non-overlapping). We know the total duration of each streak.
  2. Experienced (real-time and partially overlapping). We know each year how long a streak has lasted, but we don’t know when it will end.

Using DJIA annual returns for 1929 through 2014 (86 years), we find that: Keep Reading

Profit Drivers of Actual Short-term Algorithmic Trading?

What drives the profitability of algorithmic long-short statistical arbitrage trading (such as pairs trading) of liquid U.S. stocks? In their September 2015 paper entitled “Performance v. Turnover: A Story by 4,000 Alphas”, Zura Kakushadze and Igor Tulchinsky examine portfolio turnover and portfolio volatility as potential net return drivers for such trading. Their data source is 4,002 randomly selected portfolios (essentially synonymous with “alphas” in their lexicon) from a substantially larger survivorship bias-free pool of real trading accounts. Position holding periods for sampled portfolios range from 0.7 to 19 trading days. The authors exclude 366 portfolios with negative performance and then remove 347 portfolios as outliers for a residual sample of 3,289 portfolios. Using daily closing prices for holdings in these portfolios over an unspecified sample period, they find that: Keep Reading

High-Frequency Technical Trading of Gold and Silver?

Does simple technical analysis based on moving averages work on high-frequency spot gold and silver trading? In their August 2015 paper entitled “Does Technical Analysis Beat the Market? – Evidence from High Frequency Trading in Gold and Silver”, Andrew Urquhart, Jonathan Batten, Brian Lucey, Frank McGroarty and Maurice Peat examine the profitability of 5-minute moving average technical analysis in the gold and silver spot markets. They consider simple moving average (SMA), exponential moving average (EMA) and weighted moving average (WMA) crossing rules. These rules buy (sell) when a fast moving average crosses above (below) a slow moving average. They start with four commonly used parameter settings, all using a fast moving average of one interval paired with a slow moving average of 50, 100, 150 or 200 intervals [(1-50), (1-100), (1-150) or (1-200)]. They then test all combinations of a fast moving average ranging from 1 to 49 intervals and a slow moving average ranging from 50 to 500 intervals, generating a total of 66,297 distinct rules. To compensate for data snooping bias, they specify in-sample and out-of-sample subperiods and test whether the most successful in-sample rules work out-of-sample. They also use bootstrapping as an additional robustness test. Using 5-minute spot gold and silver prices during January 2008 through mid-September 2014, they find that: Keep Reading

SMA Signal Effectiveness Across Stock ETFs

Simple moving averages (SMA) are perhaps the most widely used and simplest market regime indicators. For example, many investors estimate that a stock index or exchange-traded fund (ETF) or individual stock priced above (below) its 200-day SMA is in a good (bad) regime. Do SMA signals/signal combinations usefully and consistently identify good and bad regimes across different kinds of U.S. stock ETFs? To investigate, we apply the regime indications of 50-day, 100-day and 200-day SMAs and some combinations of these SMAs to a variety of broad equity market (DIASPYIWBIWM and QQQ), equity style (IWDIWFIWN and IWO) and equity sector (XLBXLEXLFXLIXLKXLPXLUXLV and XLY) ETFs. As an ancillary test, we also consider three individual stocks: Apple (AAPL), Bershire Hathaway (BRKB) and Wal-Mart (WMT). Using daily dividend-adjusted closes of these 18 ETFs and three stocks from the end of July 2000 (limited by data availability for IWN and IWO) through August 2015 (about 15 years), we find that: Keep Reading

Optimal Cycle for Monthly SMA Signals?

A reader commented and asked:

“Some have suggested that the end-of-the-month effect benefits monthly simple moving average strategies that trade on the last day of the month. Is there an optimal day of the month for long-term SMA calculation and does the end-of-the-month effect explain the optimal day?”

To investigate, we compare 21 variations of a 10-month simple moving average (SMA10) timing strategy based on shifting the monthly return calculation cycle relative to trading days from the end of the month (EOM) and applied to SPDR S&P 500 (SPY) as a tradable proxy for the U.S. stock market. Using daily dividend-adjusted and unadjusted closes for SPY from inception (end of January 1993) through August 2015 and contemporaneous three-month Treasury bill (T-bill) yields, we find that: Keep Reading

Page 1 of 2212345678910...Last »
Current Momentum Winners

ETF Momentum Signal
for December 2015 (Final)

Winner ETF

Second Place ETF

Third Place ETF

Gross Compound Annual Growth Rates
(Since August 2006)
Top 1 ETF Top 2 ETFs
12.0% 12.3%
Top 3 ETFs SPY
12.6% 7.4%
Strategy Overview
Current Value Allocations

ETF Value Signal
for December 2015 (Preliminary)





The asset with the highest allocation is the holding of the Best Value strategy.
Gross Compound Annual Growth Rates
(Since September 2002)
Best Value Weighted 60-40
12.7% 9.8% 7.9%
Strategy Overview
Recent Research
Popular Posts
Popular Subscriber-Only Posts