ABSTRACT This paper develops a simple technique that controls for “false discoveries,” or mutual funds exhibit significant alphas by luck alone. Our approach precisely separates into (1) unskilled, (2) zero‐alpha, and (3) skilled funds, even with dependencies in cross‐fund estimated alphas. We find 75% of zero alpha (net expenses), consistent the Berk Green equilibrium. Further, we proportion (positive alpha) prior to 1996, but almost none 2006. also show controlling false discoveries...

We develop an econometric methodology to infer the path of risk premia from a large unbalanced panel individual stock returns. estimate time-varying implied by conditional linear asset pricing models where conditioning includes both instruments common all assets and asset-specific instruments. The estimator uses simple weighted two-pass cross-sectional regressions, we show its consistency asymptotic normality under increasing time series dimensions. address consistent estimation variance...

We consider a nonparametric method to estimate the expected shortfall—that is, loss on portfolio of financial assets knowing that is larger than given quantile. derive asymptotic properties kernel estimators shortfall and its first‐order derivative with respect allocation in context stationary process satisfying strong mixing conditions. An empirical illustration for stocks. Another deals data fire insurance losses.

We consider a nonparametric method to estimate copulas, ie, functions linking joint distributions their univariate margins. derive the asymptotic properties of kernel estimators copulas and derivatives in context multivariate stationary process satisfying strong mixing conditions. Monte Carlo results are reported for vector autoregressive order one with Gaussian innovations. An empirical illustration containing comparison independent, comotonic is given European US stock index returns.

Abstract This paper introduces two new nonparametric estimators for probability density functions which have support on the non-negative real line. These kernel are based some inverse Gaussian (IG) and reciprocal (RIG) used as kernels. We show that they share same properties those of gamma estimators: free boundary bias, always achieve optimal rate convergence mean integrated squared error (MISE). Monte Carlo results concerning finite sample reported different distributions sizes. Keywords:...

Applying tests for jumps to financial data sets can lead an important number of spurious detections. Bursts volatility are often incorrectly identified as when the sampling is too sparse. At a higher frequency, methods robust microstructure noise required. We argue that whatever jump detection test and large detections remain because multiple testing issues. propose formal treatment based on explicit thresholding available statistics. prove our method eliminates asymptotically all remaining...

We aim at accommodating the existing affine jump‐diffusion and quadratic models under same roof, namely linear‐quadratic (LQJD) class. give a complete characterization of dynamics this class by stating explicitly structural constraints, as well admissibility conditions. This allows us to carry out specification analysis for three‐factor LQJD models. compute standard transform state vector relevant asset pricing up system ordinary differential equations. show that can be embedded into using...

We consider asymmetric kernel density estimators and smoothed histograms when the unknown probability function f is defined on [0,+∞). Uniform weak consistency each compact set in [0,+∞) proved for these continuous its support. Weak convergence L1 also established. further prove that estimator histogram converge to infinity at x = 0 unbounded 0. Monte Carlo results an empirical study of shape a highly skewed income distribution based large microdata are finally provided.We thank O. Linton...

Journal Article Approximation and Calibration of Short-Term Implied Volatilities Under Jump-Diffusion Stochastic Volatility Get access Alexey Medvedev, Medvedev HEC Genève Swiss Finance Institute, Université de Address correspondence to Olivier Scaillet, Genève, UNI MAIL, Faculté des SES, 102 Bd Carl Vogt, 1211 Geneva 4, Switzerland, or e-mail: scaillet@hec.unige.ch. Search for other works by this author on: Oxford Academic Google Scholar Scaillet The Review Financial Studies, Volume 20,...

We consider consistent tests for stochastic dominance efficiency at any order of a given portfolio with respect to all possible portfolios constructed from set assets. justify block bootstrap approaches achieve valid inference in time series setting. The test statistics are computed using linear and mixed integer programming formulations. Monte Carlo results show that the procedure performs well finite samples. empirical application reveals Fama French market is first second-order efficient,...

We use the database leak of Mt. Gox exchange to analyze dynamics price bitcoin from June 2011 November 2013. This gives us a rare opportunity study an emerging retail-focused, highly speculative and unregulated market with trader identifiers at tick transaction level. Jumps are frequent events they cluster in time. The order flow imbalance preponderance aggressive traders, as well widening bid-ask spread predict them. have short-term positive impact on activity illiquidity see persistent...

We use the database leak of Mt. Gox exchange to analyze dynamics price bitcoin from June 2011 November 2013. This gives us a rare opportunity study an emerging retail-focused, highly speculative and unregulated market with trader identifiers at tick transaction level. Jumps are frequent events they cluster in time. The order flow imbalance preponderance aggressive traders, as well widening bid-ask spread predict them. have short-term positive impact on activity illiquidity induce persistent...

ABSTRACT We develop a flexible and bias‐adjusted approach to jointly examine skill, scalability, value‐added across individual funds. find that skill scalability (i) vary substantially funds, (ii) are strongly related, as great investment ideas difficult scale up. The combination of produces is positive for the majority approaches its optimal level after an adjustment period (possibly due investor learning). These results consistent with theoretical models in which funds skilled able extract...

The author considers a consistent, Kolmogorov-Smirnov type of test the complete set restrictions that relate to copula representation positive quadrant dependence. For such test, he proposes and justifies inference relying on simulation-based multiplier method bootstrap method. He also explores finite-sample behaviour both methods with Monte Carlo experiments. A first empirical illustration is given for American insurance claim data. second one examines presence dependence in life...

We consider a nonparametric method to estimate conditional expected shortfalls, i.e. losses knowing that are larger than given loss quantile. derive the asymptotic properties of kernel estimators shortfalls in context stationary process satisfying strong mixing conditions. An empirical illustration is for several stock index returns, namely CAC40, DAX30, S&P500, DJ1, and Nikkei225.