Quantitative Investors Research Articles

New accounting standards and the performance of quantitative investors

We examine quantitative investors’ ability to navigate a common and occasionally material change to the financial data generating process: new accounting standards. Returns of quantitative mutual funds temporarily decrease relative to funds that rely more heavily on human discretion following the implementation of a few standards that significantly change key financial statement variables; however, other standards do not appear to have a differential effect. Our result is stronger for quantitative funds using more accounting terminology in their prospectuses and using value strategies, which leverage accounting signals. Excess portfolio turnover following the implementation of accounting standards appears to be a driving factor of the quant underperformance. Additional evidence connects the fund-level results to the specific stocks that were affected by the accounting standards. Overall, our results suggest quant funds are generally proficient at adapting to accounting changes, although material changes occasionally put them at a temporary disadvantage relative to discretionary investors.

Distinguishing between recurring and nonrecurring components of earnings using unobserved components modeling

Distinguishing between recurring and nonrecurring components of earnings is a critical task in financial analysis and valuation. Academics and quantitative investors often rely on measures of recurring and nonrecurring components derived from standardized financial databases. We use unobserved components modeling and the Kalman smoother to obtain efficient ex-post estimates of the recurring and nonrecurring components of annual earnings. We then show that popular measures are significantly misspecified and that investors appear to anticipate a significant portion of the misspecification. Finally, we identify certain misclassified items that drive misspecification and provide algorithms to improve their ex-ante classification.

Application of machine learning in stock selection

Abstract With the development of artificial intelligence technology, machine learning has achieved very good results in the field of stock selection. This paper mainly studies the application of linear model, clustering, support vector machine, random forest, neural network and deep learning methods in the field of stock selection. The main contribution of this paper is to provide a new idea for traditional quantitative investors, so that they can build a more efficient stock selection model in practical application. The experimental results show that the stock selection model constructed by these six machine learning methods can obtain higher return and stability.

Forecasting oil futures realized range‐based volatility with jumps, leverage effect, and regime switching: New evidence from MIDAS models

AbstractThis study adds ample evidences on forecasting oil futures realized range‐based volatility (RRV) in the mixed data sampling (MIDAS) framework. Considering the features and frequent extreme risk in oil futures trading in practice, we investigate the effect of jumps, leverage effect, and regime switching. The results highlight the great importance of combing jumps, leverage effect, and regime switching simultaneously in oil futures RRV forecasting, which still holds the best predictability during a highly volatile state. Moreover, considering regime switching can also improve forecast accuracy for longer forecasting horizons such as weekly, biweekly, and monthly. The conclusions are robust to various settings and criteria. Our findings are essential for oil enterprises and quantitative investors to have a good command of crude oil futures market characteristics and achieve risk aversion.

New Accounting Standards and the Performance of Quantitative Investors

New Accounting Standards and the Performance of Quantitative Investors

Man vs. Machine: Quantitative and Discretionary Equity Management

In modern asset markets, man and machine compete for profits. How does each fare? I build a learning model in which quantitative investors (reliant on computer models) have more learning capacity but less flexibility to adapt to market conditions than discretionary investors (reliant on human judgment). I use machine learning to categorize US active equity mutual funds as quantitative or discretionary. Consistent with the model's predictions, I find that quantitative funds hold more stocks, specialize in stock picking, and engage in more overcrowded trades. Discretionary funds hold lesser known stocks, switch between picking and timing and outperform in recessions.

Is Your Alpha Pure and Your Beta Fit for Purpose?

Quantitative investors produce high backtested alpha, however, sometimes “less is more”. A beta that is unfit for purpose will favor factors with inflated alpha putting investment models at risk. Purifying alpha does not maximize past returns. It reduces unwanted risks and increases the probability of aligning live performance with backtests. Because there is no statistical procedure to choose one beta over another, the literature is filled with “better betas”. We suggest alpha purity should be the key differentiator between betas and we study how to bring alpha closer to normality. Practitioners fearful of blindly maximizing backtests will appreciate this approach.

Predicting stock market movements using network science: an information theoretic approach

A stock market is considered as one of the highly complex systems, which consists of many components whose prices move up and down without having a clear pattern. The complex nature of a stock market challenges us on making a reliable prediction of its future movements. In this paper, we aim at building a new method to forecast the future movements of Standard & Poor’s 500 Index (S&P 500) by constructing time-series complex networks of S&P 500 underlying companies by connecting them with links whose weights are given by the mutual information of 60-min price movements of the pairs of the companies with the consecutive 5340 min price records. We showed that the changes in the strength distributions of the networks provide an important information on the network’s future movements. We built several metrics using the strength distributions and network measurements such as centrality, and we combined the best two predictors by performing a linear combination. We found that the combined predictor and the changes in S&P 500 show a quadratic relationship, and it allows us to predict the amplitude of the one step future change in S&P 500. The result showed significant fluctuations in S&P 500 Index when the combined predictor was high. In terms of making the actual index predictions, we built ARIMA models with and without inclusion of network measurements, and compared the predictive power of them. We found that adding the network measurements into the ARIMA models improves the model accuracy. These findings are useful for financial market policy makers as an indicator based on which they can interfere with the markets before the markets make a drastic change, and for quantitative investors to improve their forecasting models.

Pairs Trading with Chinese Index Futures and its Implied Timing Ability

The strong government intervention and irrational investment behavior from large group of individual investors have posted challenge to quantitative investors. Many have given up their research endeavor and some even claimed Chinese stock market is a dead land for quantitative investment. We agree that some restrictions in Chinese stock market have limited the use of a few strategies. From the pure systematic trading point of view, as long as liquidity exists in Chinese future market, professional quantitative investors should be able to find appropriate strategies that work in this unique market.

Merton's problem for an investor with a benchmark in a Barndorff-Nielsen and Shephard market.

To try to outperform an externally given benchmark with known weights is the most common equity mandate in the financial industry. For quantitative investors, this task is predominantly approached by optimizing their portfolios consecutively over short time horizons with one-period models. We seek in this paper to provide a theoretical justification to this practice when the underlying market is of Barndorff-Nielsen and Shephard type. This is done by verifying that an investor who seeks to maximize her expected terminal exponential utility of wealth in excess of her benchmark will in fact use an optimal portfolio equivalent to the one-period Markowitz mean-variance problem in continuum under the corresponding Black-Scholes market. Further, we can represent the solution to the optimization problem as in Feynman-Kac form. Hence, the problem, and its solution, is analogous to Merton’s classical portfolio problem, with the main difference that Merton maximizes expected utility of terminal wealth, not wealth in excess of a benchmark.

Risk Management of Alpha Models

After posting good performance and impressive business growth for over two decades, quantitative equity investment managers have recently produced weak returns. We develop a measure of risk and show how changes in risk provide a common framework to explain past under-performance, as well as improve forecasts of factor returns. We find that quantitative stock ranking models based upon factor weights that vary with their conditional (on risk) forecasted returns are superior to traditional models with fixed weights based upon unconditional historical averages. The suggested improvements to investment processes rely upon objective and well-defined relationships between factor returns and risk, hence justifying our title - risk management of alpha models. Quantitative investors should derive comfort from this research indicating that well-defined modifications to traditional quantitative processes can go a long way towards improving returns and Sharpe ratios as well as in mitigating under-performance.

The Parallel Universes of Institutional Investing and Institutional Voting

Our paper first examines the reasons for the almost complete separation between the investment professionals at institutional investors who decide whether and when to buy or sell a company’s stock (and their counterparts at quantitative investors who design and implement the investment models), on the one hand, and the people who make voting decisions for the portfolio stocks owned by the institutional investors, on the other hand. The paper next observes the manner in which those delegated to vote portfolio stocks cope with their institution’s fiduciary duty to vote all portfolio stocks on all matters submitted to shareholders (which, depending on the diversification of an institution’s portfolio can amount literally to hundreds or thousands of votes every proxy season) in accordance with the traditional fiduciary duty of care, including the outsourcing of some or all of these duties to third party proxy advisory firms. It goes on to note that most frequently the challenge posed by the large, if not overwhelming, number of votes that need to be cast annually is met through creation and application of voting policies, based solely on concepts of corporate governance, that are applied uniformly to all corporations, regardless of the particular facts and circumstances of the different companies. The Commentary concludes by identifying some of the important implications of the parallel universes of investment decision making and voting decision making, which include: the necessity for public companies to develop communication strategies for the two very different constituencies, one that cares almost exclusively about corporate performance and the other that is concerned almost as exclusively with corporate governance practices; the desirability of establishing an inter-active and meaningful dialog among public companies and corporate governance driven voting decision makers; and, perhaps most important, the threat posed by the parallel universes to basic premises underlying corporate democracy as a foundation for the legitimacy of the modern publicly held corporation.

Beyond the Central Tendency:Quantile Regression as a Tool in Quantitative Investing

Quantitative investors frequently analyze factor performance using regression based on the familiar ordinary least squares approach. This is highly effective for understanding the central tendency within a dataset, but will often be less useful for assessing the behavior of datapoints close to the upper or lower extremes within a population. But from the perspective of active investors or risk managers, the datapoints at the extremes may be precisely the ones of greatest interest. For such applications, a more appropriate methodology is quantile regression. The authors show how quantile regression represents an extension of the conventional ordinary least squares method, and present an empirical analysis of factor effectiveness applied to a universe of U.S. small-cap stocks in order to illustrate the insights offered by this technique. <b>TOPICS:</b>Portfolio construction, statistical methods, accounting and ratio analysis