Multiple hypothesis testing is a fundamental problem in high dimensional inference, with wide applications in many scientific fields. In genome-wide as-sociation studies, tens of thousands of tests are performed simultaneously to find if any SNPs are associated with some traits and those tests are correlated. When test statistics are correlated, false discovery control becomes very chal-lenging under arbitrary dependence. In the current paper, we propose a novel method based on principal factor approximation, which successfully subtracts the common dependence and weakens significantly the correlation structure, to deal with an arbitrary dependence structure. We derive an approximate expression for false discovery proportion (FDP) in large scale multiple testing when a common threshold is used and provide a consistent estimate of realized FDP. This result has important applications in controlling FDR and FDP. Our estimate of realized FDP compares favorably with Efron (2007)’s approach, as demonstrated in the simulated examples. Our approach is further illustrated by

Mutual funds change their risk levels significantly over time. Risk shifting might be caused by ill-motivated trades of unskilled or agency-prone fund managers who trade to increase their personal compensation. Alternatively, risk shifting might occur when skilled fund managers trade to take advantage of their stock selection and timing abilities. This paper investigates the performance consequences of risk shifting and sheds light on the mechanisms and the economic motivations behind the risk shifting behavior. Using a holdings-based measure of risk shifting, we find that funds that increase risk perform worse than funds that keep stable risk levels over time, suggesting that risk shifting is either an indication of inferior ability or is motivated by agency issues.

Abstract not found

Multiple hypothesis testing is a fundamental problem in high dimensional inference, with wide applications in many scientific fields. In genome-wide as-sociation studies, tens of thousands of tests are performed simultaneously to find if any genes are associated with some traits and those tests are corre-lated. When test statistics are correlated, false discovery control becomes very challenging under arbitrary dependence. In the current paper, we propose a new methodology based on principal factor approximation, which successfully substracts the common dependence and weakens significantly the correlation structure, to deal with an arbitrary dependence structure. We derive the theo-retical distribution for false discovery proportion (FDP) in large scale multiple testing when a common threshold is used and provide a consistent FDP. This result has important applications in controlling FDR and FDP. Our estimate of FDP compares favorably with Efron (2007)’s approach, as demonstrated by in the simulated examples. Our approach is further illustrated by some real data applications.

a b s t r a c t We revisit the apparent historical success of technical trading rules on daily prices of the Dow Jones Industrial Average index from 1897 to 2011, and we use the false discovery rate (FDR) as a new approach to data snooping. The advantage of the FDR over existing methods is that it selects more outperforming rules, which allows diversifying against model uncertainty. Persistence tests show that, even with the more powerful FDR technique, an investor would never have been able to select ex ante the future best-performing rules. Moreover, even in-sample, the performance is completely offset by the introduction of low transaction costs. Overall, our results seriously call into question the economic value of technical trading rules that has been reported for early periods. & 2012 Elsevier B.V. All rights reserved. Introduction Whether technical trading rules can consistently generate profits, as opposed to just being lucky every now and then, is the subject of an ongoing debate. Practitioners have devoted significant resources to technical trading, which uses past price and volume data to infer future prices. A substantial segment of the investment industry employs indicators that include moving averages, support and resistance levels, and other filter rules. Technical indicators are as ubiquitous on professional information systems as on popular finance websites and online retail brokers. In spite of its popularity among practitioners, academics have long been skeptical about the merits of technical analysis. $ We would like to thank the editor and the referee for constructive criticism and numerous suggestions that have lead to substantial improvements over previous versions of the paper. We are grateful to M. Franscini-Scaillet for helping us to get the data on fund structure costs and futures trading costs via her industry contacts. We thank L. Barras, I.

doi 10.1287/mksc.1090.0505

Abstract We examine commercial real estate fund managers' abilities to generate abnormal profits through selection of outperforming property sub-market segments or through the timing of entry into and exit from sub-markets. The vast majority of portfolio managers exhibit little market timing ability, with the exception of non-NYSE REITs after the financial crisis. A substantial fraction of managers seem able to successfully select property sub-markets. Selection performance exhibits significant persistence. Managers that are active in more liquid markets tend to exhibit better timing performance, while managers exhibiting better selection ability appear to be active in less liquid markets.

We empirically analyze the nature of returns to scale in active mutual fund management. We find strong evidence of decreasing returns at the industry level: As the size of the active mutual fund industry increases, a fund’s ability to outperform passive benchmarks typically declines. In contrast, estimates that avoid econometric biases do not find decreasing returns at the fund level. We also find that funds born more recently exhibit more skill. This upward trend in skill coincides with industry growth, which precludes the skill improvement from boosting average fund performance. Finally, we find that a fund’s performance typically declines over its lifetime. This striking result can also be explained by industry growth and industry-level decreasing returns to scale.

Mutual funds often disappear following poor performance. When this poor perfor-mance is partly attributable to negative idiosyncratic shocks, funds ’ estimated alphas understate their true alphas. This paper estimates a structural model to correct for this bias. Although most funds still have negative alphas, they are not nearly as low as those suggested by the fund-by-fund regressions. Approximately 12 % of funds have net four-factor model alphas greater than 2 % per year. All studies that run fund-by-fund regressions to draw inferences about the prevalence of skill among mutual fund managers are subject to reverse survivorship bias. THE TYPICAL SURVIVORSHIP BIAS argument starts from the observation that mutual funds often disappear following poor performance. Thus, a study that conditions on fund survival overstates performance.1 In this paper, I show that the correlation between performance and survival induces an-other bias with the opposite sign: mutual fund alphas estimated from a sur-vivorship bias-free data set are biased downward relative to the true alpha

This paper re-examines the incentives of mutual fund managers arising from investor flows. We provide evidence that the convexity of the flow-performance relationship varies with economic activity. We show that the effect is economically large and is not driven by abnormal years. We test two possible channels through which this pattern may arise. We investigate implications of the timevarying convexity for the incentives of managers to alter strategically the risk of their portfolios. We provide evidence that poor mid-year performers increase the risk of the portfolio only when economic activity is strong. Finally, we briefly discuss some methodological implications.