New strategies:
#314 – Sector Rotation Strategy Based on Multivariate Regression Analysis
Period of rebalancing: monthly
Markets traded: equities
Instruments used for trading: ETFs
Complexity: Complex strategy
Bactest period: 2000-2015
Indicative performance: 7.04%
Estimated volatility: 10.06%
Source paper:
Seggebruch: Multivariate Regression Analysis: Considering the Relevance of Past Performance
http://artesysonline.com/wp-content/uploads/2016/06/Multivariate-Regression-Analysis-Considering-the-Relevance-of-Past-Performance.pdf
Abstract:
Investors often use past performance as a major source of knowledge about an asset class or a particular investment manager. Past performance can tell us a lot about the tendencies of asset classes and managers, but its meaning should be evaluated with great care. Simply comparing performances over an arbitrary time period can give way to a pattern of return chasing that can severely detract from performance. In fact, the emotional behavior of changing investments based on past returns has been the topic of any publications and hours of research. This paper will provide a statistical perspective on the relevance of past performance. Specifically, I will show that multivariate regression analysis can successfully identify mathematical relationships between various past performance statistics and future returns.
Multivariate regression analysis is not a foreign concept to the financial industry. It has been utilized by technical and quantitative analysts for some time now. However, constructing and interpreting this type of statistical analysis can be obstacles to investors without technical backgrounds. With this in mind, I will give a description of each of the variables and results in the analysis and inform the reader of what is necessary for a statistically significant prediction model. I believe that this insight will make the benefits of multivariate regression analysis accessible to a wider variety of investors.
I begin with a concise description of the assets used as representation for the sectors of the S&P 500. This description is followed by an explanation of the method used for calculating returns and the frequency with which they are calculated. These two variables, calculation and frequency, are often debated topics and can have material effects on the outcome of the analyses. In this paper, I will use a monthly return frequency to calculate a logarithmic return. These decisions are key to multivariate regression analysis and must be made before further analysis is completed, making comparison between the outcomes of using different frequencies and return calculations quite time-consuming. However, it does force this decision to be a forethought and lessens the bias that could be present if it were an afterthought.
Investors often select calculation time periods with hindsight bias by comparing di erent returns and selecting the one that looks the best at that point in time. To better answer the questions, "What time period should be used for performance calculations?" And, "How long is the prediction good for?" I will attach statistical signi cance to four different look back time periods and four different future time periods and make an informed decision on which combination has the highest predictive capability. By modeling each of these different combinations of time periods, we gain insight about the sensitivity of the analysis to the time period variable. I will show that varying levels of significance exists across the different time period combinations and select one pair to be the optimal time period combination. I will discuss results from each look back analysis, but for brevity this will not be an exhaustive exposition.
Finally, I will display the predictive capability of employing a multi-variate regression model from the optimal time period combination in an actively managed sector rotation trading system. The performance that results from this trading system outperforms the S&P 500 on a risk adjusted basis according to several well-accepted performance measures. This success is mainly due to the down side protection incurred by rotating through the US equity sectors via a rules-based decision making process, while still participating on the up side. For further analytical rigor, I forward tested the trading strategy 36 months to verify the analysis with out-of-sample data. I will show that the trends discovered by multivariate regression analysis are also present in data excluded from the backtest. Ultimately, I will show that a rules-based trading system, supplemented with multivariate regression, is a viable alternative to investing in a passive index.
#315 – Stock Splits Strategy Based on Earnings Management
Period of rebalancing: yearly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Complex strategy
Bactest period: 1980-2011
Indicative performance: 11.35%
Estimated volatility: not stated
Source paper:
Elnahas, Gao and Ismail: Long-Term Returns Predictability Following Stock Splits: The Blind Side
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2800772
Abstract:
Prior studies suggest that combining stock split signal and accrual signal is an effective means of communicating managerial optimism, and that stock split is an exception to the accruals anomaly. We posit that, although stock splits are mostly conducted by optimistic managers and are followed by positive long-term returns, many are conducted by overoptimistic and/or opportunistic managers and are followed by negative long-term returns. Disentangling optimistic and overoptimistic/opportunistic managers helps ex-ante identify bad splits that are followed by significantly negative abnormal returns. A zero-investment strategy based on our findings can generate economically and statistically significant positive returns.
New research paper related to existing strategy:
#1 – Asset Class Trend Following
Faber: The Trinity Portfolio: A Long-Term Investing Framework Engineered for Simplicity, Safety, and Outperformance
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2801856
Abstract:
Let’s say one sets out to design a portfolio, knowing everything we know today about investing. Where would a logical, evidence-based investor even start? Investors today have access to more market data and strategic information than at any other time in history. While beneficial in some ways, this huge volume of fragmented information presents a challenge — how should one actually implement everything? This paper offers a potential solution – the “Trinity Portfolio.” The name is a reference to the three core elements of the portfolio: 1) assets diversified across a global investment set, 2) tilts toward investments exhibiting value and momentum traits, and 3) exposure to trend following. We examine how an investor might construct this holistic, adaptive framework consisting of some of the most well-known market anomalies. We find that the portfolio performs well across various market environments, with reasonable volatility. Finally, we examine how an investor may update and implement such a portfolio with low cost funds.
Two additional related research papers have been included into existing free strategy reviews during last 2 week:
Explanatory research paper for all short term contrarian strategies:
Kim: Black Monday, Globalization and Trading Behavior of Stock Investors
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2798536
Abstract:
Using a simple sign test, we report new empirical evidence, taken from both the US and the German stock markets, showing that trading behavior substantially changed around Black Monday in 1987. It turned out that before Black Monday investors behaved more as in the momentum strategy; and after Black Monday more as in the contrarian strategy. We argue that crashes, in general, themselves are merely a manifestation of uncertainty on stock markets and the high uncertainty due to globalization is mainly responsible for this change.
#14 – Momentum Effect in Stocks
Bhattacharya, Li, Sonaer: Has Momentum Lost Its Momentum?
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2791138
Abstract:
We evaluate the robustness of momentum returns in the US stock market over the period 1965 to 2012. We find that momentum profits have become insignificant since the late 1990s partially driven by pronounced increase in the volatility of momentum profits in the last 14 years. Investigations of momentum profits in high and low volatility months address the concerns about unprecedented levels of market volatility in this period rendering momentum strategy unprofitable. Past returns, can no longer explain the cross-sectional variation in stock returns, even following up markets. Investigation of post holding period returns of momentum portfolios and risk adjusted buy and hold returns of stocks in momentum suggests that investors possibly recognize that momentum strategy is profitable and trade in ways that arbitrage away such profits. These findings are partially consistent with Schwert (2003) that documents two primary reasons for the disappearance of an anomaly in the behavior of asset prices, first, sample selection bias, and second, uncovering of anomaly by investors who trade in the assets to arbitrage it away. In further analyses we find evidence that suggest three possible explanations for the declining momentum profits that involve uncovering of the anomaly by investors, decline in the risk premium on a macroeconomic factor, growth rate in industrial production in particular and relative improvement in market efficiency.
