Tag Archives: paper

A Pioneering Approach To Earnings Season

Summary US Q2 earnings season was one of the most volatile ever. We present Pioneering Quantitative Approach focusing on prices and not on fundamental data. Our analysis provides a guide per sector and per capitalization. ABSTRACT “Qui sait le passé peut conjecturer l’avenir”, Jacques-Bénigne Bossuet As we are just about to enter in the Q3 Earnings Season in the US, Uncia AM decided to provide you some keys to understand the Q2 Earnings Season. Bloomb erg already gave some keys: US stocks got biggest earnings bang since 2012 . This empirical study emphasizes many things: 1/ It is not worthwhile to keep equity position over the earnings: earnings releases are a lottery. As difficult as it may be for our equity analyst friends to admit (note: the author is an asset manager) , all available empirical data shows that it is impossible to predict market reaction following an earnings release. We thus need to distinguish the fundamental component of the reaction which is less unpredictable (related to turnover, EBITDA and other hard data) from the “price signal” component. The latter has always been impossible to predict, even if we take into account the released fundamental data. From a statistical perspective, the specific movement linked an the earnings release is on average null, as can be seen from the highly leptokurtic distribution of the movement. For an asset manager seeking to optimize their Sharpe Ratio, it is therefore not worth maintaining a position in the equity over the release period (assuming transaction & liquidity fees to be marginal) (click to enlarge) Source: Bloomberg, Uncia AM, Alphametry. Read the entire article in order to make your own opinion : 2/ Information Technology sector behaved properly during this session, on almost all indices. 3/ The specific Russell 2000 – related stocks moved a lot on earnings: maybe more interest of investors for UScentric names, as a consequence of fear over the USD strength and the world/Chinese macro slowdown. As the article may be a bit technical, here is a brief takeaway: On average, stocks from Nasdaq Composite Index (NASDAQ: CCMP ) exhibit a null return over earnings, but with large volatility. Therefore, it justifies the strategy to cut positions over earnings. In addition to that, we can notice that signals were slightly better on large caps vs small caps, and quite good in a sector such as Information Technology. “Weekly speaking”, we experienced a sharp positive signal on CCMP, but on a “PEAD” perspective only few comonotony between Earnings Moves and Drift Returns. On average, stocks from Russell 2000 Index (RTY) exhibit a null return over earnings, but with large volatility. Therefore, it justifies the strategy to cut positions over earnings. In addition to that, we can notice that signals were slightly better on large caps vs small caps, and quite good on a sector such as Information Technology, same things as we notice on CCMP. There are a lot more “PEAD” signals on RTY than on CCMP, meaning that as there are many companies belonging to both indices, many companies belonging only to RTY exhibit large signals. This means that investor attention was largely focused on UScentric companies. “Weekly speaking”, since the beginning of the year, we had very positive signals on RTY, but the summer was very complicated as we can see a downside candle at the beginning of August. Stocks from S&P 500 (SPX) are less volatile over earnings than those of CCMP, RTY or Nasdaq 100 (NDX). It may be explained by the average size of capitalization, but this is not sufficient as NDX average capitalization is higher (58.0 vs 50.2) is higher than SPX. We make the same notification about earnings release volatility that is not rewarded, unless capitalization criteria is not worthwhile anymore, nor sector criteria (even if we can see a positive skew for Information Technology sector). In terms of “Weekly signals”, we can notice numerous negative signals, emphasizing an overreaction of investors about bad news versus good news. We have only few data, but first of all, we can notice that stocks from Nasdaq 100 (NDX) exhibits the largest average capitalization, and the largest absolute earnings moves. For more technical readers, should you be interested in the underlying philosophy, please go ahead: METHODOLOGY Our sample takes into account earnings that occurred between 2015, June 30th and 2015, August 31st. We only focus on companies whose market capitalization exceeds $1 billion, the day before the earnings release (ER)/call (NYSE: EC ). We focus on 4 main US indices: Nasdaq Composite , Nasdaq 100 (NDX), S&P 500 (SPX) and Russell 2000 (RTY). Our method to estimate the move due to earnings release/call is the following: We assume that the Management Call lasts one hour, and that ER had occurred just before, which is the standard case (hugely often- we consider it happens all the time). Therefore, thinking as of Paris time, with 6-hours delay with New-York, we can set the following table: Table of earnings category Source: Uncia AM. We use the earnings return by getting rid off the total return index to the idiosyncratic move, assuming a beta for each stock = 1. For more information, you can refer to the original paper by the author, Post Earnings Announcement Drift, a Price Signal? [1] Important: in the following development, return always refers to relative return of the stock versus its index (total-return). NASDAQ COMPOSITE – CCMP: average capitalization (

Combining Volatility, Momentum, And Trend In Asset Allocation

Summary Risk-based portfolio strategies are popular in the asset management industry. Three common strategies are Minimum Variance (MV), Equal Risk Contribution (ERC) and Maximum Diversification (MD). These strategies do not depend on asset returns’ forecasts and they are based on a single criterion: risk. With higher returns and lower risk, risk-based portfolios that use moving averages have higher Sharpe ratios than initial risk-based portfolios. High momentum risk-based portfolios, by contrast, have higher risk, which is largely compensated for by higher returns. By Gregory Guilmin The Effective Combination of Risk-Based Strategies with Momentum and Trend Following Abstract: The Efficient Market Hypothesis (EMH) has been widely called into question in the investment literature, through two main anomalies: timing and low-volatility anomalies. In this paper, we aim to combine the predictive power of timing and low-volatility strategies to deliver better risk-adjusted portfolio performance. We adopt a two-step approach for a constant dataset composed of 18 country MSCI stock market indices over the 1975-2014 period. First, we use different timing strategies: moving averages and momentum. We select stock market indices based on different moving averages (6, 8, 10, and 12 months), while the momentum strategy ranks the different stock market indices into momentum subsets (low, medium, and high momentum). After the first step using the different timing strategies, the second step consists in building risk-based portfolios (MV, ERC, and MD) as well as 1/N benchmark portfolios for each of these timing strategies. Our results highlight the effectiveness, the relevance and the robustness of our approach. First, risk-based portfolios using relevant timing strategy indeed provide better returns, lower volatilities, higher Sharpe ratios, and lower Value-at-Risk (VaR) and Expected Shortfall (ES) than traditional risk-based portfolios. The second contribution of our approach features that risk-based strategies provide better risk-adjusted returns and lower VaR and ES than the 1/N portfolio within a context in which the first step is dedicated to the application of a relevant timing strategy. Finally, among these risk-based portfolios using relevant timing strategies, the MD and MV portfolios usually obtain the best risk-adjusted performance. Alpha Highlight: Risk-based portfolio strategies are popular in the asset management industry. Three common strategies are Minimum Variance (MV), Equal Risk Contribution (ERC) and Maximum Diversification (MD). These strategies do not depend on asset returns’ forecasts and they are based on a single criterion: risk. The interest in estimation procedures relying on a risk measure could be explained by three major factors: Reconsideration of the importance and the relevance of portfolio risk management. Better predictability of security variance and covariance risks by comparison with expected returns. The outperformance of the “low-volatility anomaly.” Details here . In addition to the low-volatility anomaly, a large number of authors have talked about the “momentum anomaly.” The momentum effect has been emphasized by Jegadeesh and Titman (1993). Momentum strategies are profitable in most major stock markets worldwide and this outperformance of momentum strategies is consistent over time. Linked to this concept of cross sectional momentum, time-series momentum, such as trend following strategies, have been identified. Several academic papers show that moving average trading rules have predictive power for future returns, and that trend following strategies with moving averages are effective in practice (Among others, see Brock et al., 1992; Clare et al., 2014; Faber, 2007, 2013; Hurst et al., 2010 and ap Gwilym et al., 2010). Methodology: Given this backdrop, in which risk-based strategies and timing strategies have been developed in the literature, the purpose of this paper is to combine the two strategies. This two-step approach consists in applying a timing strategy (either a moving average or a momentum strategy in the first step) followed by risk-based portfolio optimization procedures (second step). We compute risk-based and equally weighted (as a benchmark) portfolios with and without timing strategies in the first step for a constant empirical dataset composed of 18 country MSCI stock market indices. The estimation period ranges from January 1975 to December 2014. To the best of our knowledge, this paper is the first to shed light on the combination of timing and risk-based strategies. First Step: Selection of the stock market indices The methodology consists of two steps. In the first step, moving averages are used to select stock market indices that perform well (by exhibiting an upward trend) and that are used in the second step of our analysis (i.e., in the risk-based and 1/N portfolio optimization). Stock market indices exhibiting a negative trend are not selected as an input in the portfolio optimization procedure. If the price of the stock market index is above its x − month moving average, then this index is selected for the portfolio optimization procedure. Conversely, if it crosses below its x − month moving average, then the stock market index is not selected for the second step. We use moving averages of varying lengths: 6, 8, 10, and 12 months. To add an additional timing strategy, we also select stock market indices in accordance with the concept of momentum, in which a stock market’s performance relative to its peers predicts its future relative performance. As long-term investors, the momentum strategy involves ranking stock market indices based on their past 12-month performance and splitting them into three subsets, depending on the value of their momentum compared with one of their peers. The three subsets are the low, medium and high momentum subsets, respectively. Second Step: Portfolio optimization After selecting stock market indices following the different timing strategies of the first step, the second stage consists in applying different portfolio optimization procedures to find the optimal weights of the selected stock market indices. Selection (1st step) and weighting (2nd step) are adjusted simultaneously, i.e., on a monthly basis (end of month). We apply three risk-based portfolio strategies (Minimum Variance, Equal Risk Contribution and Maximum Diversification) as well as the 1/N benchmark portfolio strategy, usually considered a relevant benchmark in the literature. First, the Minimum Variance portfolio aspires to minimize the global variance of the portfolio. Second, the Equal Risk Contribution portfolio is the portfolio in which the risk contribution is the same for all assets in the portfolio. Finally, the Maximum Diversification portfolio (also called the Most Diversified Portfolio), introduced by Choueifaty (2006), is the portfolio that maximizes diversification. Diversification is computed using the diversification ratio. The diversification ratio is defined as the ratio of its weighted average volatility to its portfolio volatility. Main Findings: The table below summarizes results based on a constant dataset composed of 18 country MSCI stock market indices between 1975 and 2014. We can see that all portfolios that employ moving averages in the first step perform better than initial risk-based portfolios . Regarding momentum, high momentum risk-based strategies offer better annual performance than initial risk-based portfolios. Table 1: Portfolio performances with the constant country MSCI Indices sample (1975-2014) Annualized Annual Annualized Sharpe VaR (1%) ES (1%) returns (%) returns (%) Volatility (%) Ratio (%) (%) Initial Portfolios 1/N 8.931 10.013 16.723 0.599 -13.118 -18.531 MV 8.152 8.866 14.017 0.632 -10.987 -14.785 ERC 8.785 9.781 16.131 0.606 -12.550 -18.116 MD 8.624 9.655 16.275 0.593 -14.254 -17.890 With timing strategies 6 months 1/N 10.032 10.732 14.965 0.717 -11.946 -15.482 MV 9.987 10.472 13.439 0.779 -10.238 -13.884 ERC 9.605 10.256 14.387 0.713 -11.592 -15.107 MD 10.515 11.166 14.882 0.750 -11.966 -15.843 8 months 1/N 10.401 11.062 14.897 0.743 -11.946 -15.363 MV 10.626 11.056 13.419 0.824 -10.212 -13.961 ERC 10.107 10.694 14.218 0.752 -11.592 -15.000 MD 11.079 11.650 14.664 0.794 -11.966 -15.787 10 months 1/N 9.832 10.534 14.842 0.710 -11.665 -15.258 MV 9.299 9.853 13.499 0.730 -10.871 -14.453 ERC 9.486 10.119 14.170 0.714 -11.311 -15.026 MD 10.191 10.855 14.744 0.736 -11.966 -16.022 12 months 1/N 10.612 11.229 14.725 0.763 -11.904 -15.220 MV 10.523 10.957 13.358 0.820 -10.602 -14.488 ERC 10.494 11.020 14.021 0.786 -11.621 -15.066 MD 11.206 11.766 14.671 0.802 -12.325 -16.304 Low momentum 1/N 5.359 7.021 18.821 0.373 -14.322 -19.843 MV 5.533 6.855 17.063 0.402 -13.632 -17.501 ERC 5.162 6.704 18.127 0.370 -14.152 -19.464 MD 5.773 7.233 17.884 0.404 -13.852 -19.023 Medium momentum 1/N 8.919 9.997 16.747 0.597 -12.766 -17.035 MV 9.245 10.101 15.602 0.647 -11.012 -15.796 ERC 9.069 10.087 16.456 0.613 -12.456 -16.953 MD 9.247 10.309 16.770 0.615 -14.402 -18.084 High momentum 1/N 11.896 13.022 18.266 0.713 -14.637 -20.915 MV 12.583 13.446 17.231 0.780 -11.539 -18.729 ERC 12.16 13.178 17.834 0.739 -13.770 -20.125 MD 12.092 13.218 18.356 0.720 -13.443 -20.719 Market-cap weighted benchmarks MSCI World 8.016 8.839 14.782 0.598 -11.073 -14.564 MSCI World Momentum 11.432 12.130 15.817 0.767 -11.517 -14.965 The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request. Trend following and high momentum strategies are effective for 1/N portfolio optimization but also for risk-based portfolios because they produce better annual returns compared with initial risk-based portfolios. With respect to risk measures such as volatility, Value-at-Risk (VaR) and Expected Shortfall (ES), risk-based portfolios that employ moving averages exhibit lower volatility than initial risk-based portfolios as well as lower VaR and ES. This finding is important because it enables investors to reduce the risk to which they are exposed. With higher returns and lower risk, risk-based portfolios that use moving averages have higher Sharpe ratios than initial risk-based portfolios. High momentum risk-based portfolios, by contrast, have higher risk, which is largely compensated for by higher returns. Therefore, such portfolios are characterized by higher Sharpe ratios than initial risk-based portfolios. This paper documents the effectiveness, in terms of risk and return, of the use of these relevant timing strategies combined with risk-based portfolio strategies. In addition to that, robustness checks were also conducted with different other datasets, different estimation periods as well as different parameters of the variance-covariance matrix. Original Post

The Active Share Debate: AQR Versus The Academics

By Jack Vogel, Ph.D. There is an interesting discussion in the geeky world of academic finance literature between the intellectual muscle at AQR and academia. The discussion revolves around the following question: ” Does Active Share matter? ” This is an important topic for active ETFs and Mutual Funds in the marketplace. The original paper on this measure was written by Cremers and Petajisto and was published in the Review of Financial Studies in 2009 (top finance journal). Links to the paper can be found here and here . The abstract of the paper is the following: We introduce a new measure of active portfolio management, Active Share, which represents the share of portfolio holdings that differ from the benchmark index holdings. We compute Active Share for domestic equity mutual funds from 1980 to 2003. We relate Active Share to fund characteristics such as size, expenses, and turnover in the cross-section, and we also examine its evolution over time. Active Share predicts fund performance : funds with the highest Active Share significantly outperform their benchmarks, both before and after expenses, and they exhibit strong performance persistence. Nonindex funds with the lowest Active Share underperform their benchmarks. Main Finding of the paper: For non-index funds, the higher the active share, the better the performance. We tend to agree, as we have talked about diworsification in the past. However, just because a manager creates a more active portfolio (a necessary condition for outperformance ), this doesn’t imply an active manager will actually have outperformance. The team at AQR (Frazzini, Friedman, and Pomorski), in a forthcoming article in the Financial Analyst Journal (link to the paper is here ), address this question. The abstract is the following: We investigate Active Share, a measure meant to determine the level of active management in investment portfolios. Using the same sample as Cremers and Petajisto (2009) and Petajisto (2013) we find that Active Share correlates with benchmark returns, but does not predict actual fund returns ; within individual benchmarks, it is as likely to correlate positively with performance as it is to correlate negatively. Our findings do not support an emphasis on Active Share as a manager selection tool or an appropriate guideline for institutional portfolios. Main point of the paper: Active share should not be used as a manager selection tool. Basically, for a given index, they find that active share cannot be used as a reliable tool to identify out-performance. So is Active Share a waste of time? As Lee Corso says every Saturday morning during College Gameday, “Not so fast!” The two authors of the original paper, Martijn Cremers and Antti Petajisto were quick to shoot down the AQR findings. Here is the executive summary from Antti Petajisto: All of the key claims of AQR’s paper were already addressed in the two cited Active Share papers: Petajisto (2013) and Cremers and Petajisto (2009). 1) The fact about the level of Active Share varying across benchmarks has been widely known for many years. Its performance impact was explicitly studied and discussed in the first drafts of Petajisto (2013) back in 2010, and the performance results remained broadly similar. The reason for the apparent discrepancy is AQR’s choice of summarizing results by benchmark, which effectively gives the same weight to the most popular index (S&P 500, assigned to 870 funds) and the least popular index (Russell 3000 Growth, assigned to 24 funds), which is not sensible as a statistical approach. 2) The issue about four-factor alphas varying across benchmark indices does nothing to change the fact that higher Active Share managers have been able to beat their benchmark indices. However, it does raise an interesting point about the four-factor approach to measuring performance, and in fact my coauthors and I wrote a long and detailed paper about this exact issue first in 2007 (published later as Cremers, Petajisto, and Zitzewitz (2013)). 3) AQR’s researchers argue that there is no theory behind Active Share and they remain mystified by the differences between Active Share and tracking error. It is unfortunate that they have entirely missed the lengthy sections of both Active Share papers that discuss this exact topic: pages 74-77 in Petajisto (2013) and sections 1.3, 3.1, and 4.1 in Cremers and Petajisto (2009). The short answer is that Active Share is more about stock selection, whereas tracking error is more about exposure to systematic risk factors. So clearly ignoring large and essential parts of the original Active Share papers is simply not the way to conduct impartial scientific inquiry. If that executive summary wasn’t scathing enough, Martijn Cremers actually wrote a paper titled ” AQR in Wonderland: Down the Rabbit Hole of ‘Deactivating Active Share’ (and Back Out Again?) ” Here is the abstract: The April 2015 paper “Deactivating Active Share”, released by AQR Capital Management, aims to debunk the claim that Active Share (a measure of active management) predicts investment performance. The claim of the AQR paper is that “neither theory nor data justify the expectation that Active Share might help investors improve their returns,” arguing that previous results are “entirely driven by the strong correlation between Active Share and the benchmark type.” This paper’s first and main aim is to establish that the AQR paper should not be interpreted using typical academic standards. Instead, our conjecture is that this AQR paper falls into a wonderfully creative but altogether different genre, which we label the Wonderland Genre, as its main characteristic seems to be “Sentence First, Verdict Later.” For example, the results in the AQR-WP cannot be taken at face value, as the information that is not shared reverses their main conclusion. Secondarily, we consider the plausible claim that benchmark styles matter and find that controlling for the main benchmark style, the predictability of Active Share is robust. While Active Share is only one tool among many to analyze investment funds and needs to be carefully interpreted for each fund individually, Active Share may indeed plausibly help investors improve their returns. Thirdly and finally, we impolitely consider why AQR may not be a big fan of Active Share by taking a look at the AQR mutual funds offered to retail investors. We find that these tend to have relatively low Active Shares, have shown little outperformance to date (with performance data ending in 2014) and thus seem fairly expensive given the amount of differentiation they offer. So who is the winner in the debate? The answer is both are probably correct at some level. More concentration (less diworsification) probably has higher active share and in the past had higher returns. However, one cannot just take any random selection of stocks and expect to outperform, the style of the investment matters, which was AQR’s argument (we prefer Value and Momentum ). Let us know what you think! Link to the original post on Alpha Architect