Analyzing the Performance of Multifactor Investment Strategies under a Multiple Testing Framework

The problem of finding periodically expressed genes from time course microarray experiments is at the center of numerous efforts to identify the molecular components of biological clocks. We present a new approach to this problem based on the cyclohedron test, which is a rank test inspired by recent advances in algebraic combinatorics. The test has the advantage of being robust to measurement errors, and can be used to ascertain the significance of top-ranked genes. We apply the test to recently published measurements of gene expression during mouse somitogenesis and find 32 genes that collectively are significant. Among these are previously identified periodic genes involved in the Notch/FGF and Wnt signaling pathways, as well as novel candidate genes that may play a role in regulating the segmentation clock. These results confirm that there are an abundance of exceptionally periodic genes expressed during somitogenesis. The emphasis of this paper is on the statistics and combinatorics that underlie the cyclohedron test and its implementation within a multiple testing framework.

Download Full-text

Local joint testing improves power and identifies missing heritability in association studies

10.1101/040089 ◽

2016 ◽

Author(s):

Brielin C. Brown ◽

Alkes L. Price ◽

Nikolaos A. Patsopoulos ◽

Noah Zaitlen

Keyword(s):

Multiple Testing ◽

Association Studies ◽

Genetic Association Studies ◽

Testing Procedure ◽

Missing Heritability ◽

Multiple Testing Correction ◽

Testing Framework ◽

Genome Wide ◽

Disease Associations ◽

Causal Variants

AbstractThere is mounting evidence that complex human phenotypes are highly polygenic, with many loci harboring multiple causal variants, yet most genetic association studies examine each SNP in isolation. While this has lead to the discovery of thousands of disease associations, discovered variants account for only a small fraction of disease heritability. Alternative multi-SNP methods have been proposed, but issues such as multiple testing correction, sensitivity to genotyping error, and optimization for the underlying genetic architectures remain. Here we describe a local joint testing procedure, complete with multiple testing correction, that leverages a genetic phenomenon we call linkage masking wherein linkage disequilibrium between SNPs hides their signal under standard association methods. We show that local joint testing on the original Wellcome Trust Case Control Consortium dataset leads to the discovery of 29% more associated loci that were later found in followup studies containing thousands of additional individuals. These loci double the heritability explained by genome-wide significant associations in the WTCCC dataset, implicating linkage masking as a novel source of missing heritability. Furthermore, we show that local joint testing in a cis-eQTL study of the gEUVADIS dataset increases the number of genes discovered by 10.7% over marginal analyses. Our multiple hypothesis correction and joint testing framework are available in a python software package called jester, available at github.com/brielin/Jester.

Download Full-text

A Bayesian Approach to Measurement of Backtest Overfitting

Risks ◽

10.3390/risks9010018 ◽

2021 ◽

Vol 9 (1) ◽

pp. 18

Author(s):

Jiří Witzany

Keyword(s):

Bayesian Approach ◽

Performance Indicators ◽

Multiple Testing ◽

Historical Data ◽

Broad Class ◽

Investment Strategies ◽

Robust Estimates ◽

Expected Performance ◽

Mcmc Simulation ◽

Naive Approach

Quantitative investment strategies are often selected from a broad class of candidate models estimated and tested on historical data. Standard statistical techniques to prevent model overfitting such as out-sample backtesting turn out to be unreliable in situations when the selection is based on results of too many models tested on the holdout sample. There is an ongoing discussion of how to estimate the probability of backtest overfitting and adjust the expected performance indicators such as the Sharpe ratio in order to reflect properly the effect of multiple testing. We propose a consistent Bayesian approach that yields the desired robust estimates on the basis of a Markov chain Monte Carlo (MCMC) simulation. The approach is tested on a class of technical trading strategies where a seemingly profitable strategy can be selected in the naïve approach.

Download Full-text

Pricing models of shares of Russian companies and their practical application

Voprosy Ekonomiki ◽

10.32609/0042-8736-2019-3-48-76 ◽

2019 ◽

pp. 48-76 ◽

Cited By ~ 1

Author(s):

Alexander E. Abramov ◽

Alexander D. Radygin ◽

Maria I. Chernova

Keyword(s):

Stock Market ◽

Institutional Investors ◽

Factor Model ◽

Investment Strategies ◽

Pricing Models ◽

Foreign Portfolio Investment ◽

Market Portfolio ◽

Stock Market Performance ◽

Long Time ◽

Russian Stock Market

The article analyzes the problems of applying stock pricing models in the Russian stock market. The novelty of the study lies in the peculiarities of the methodology used and the substantive conclusions on the specifics of the influence of fundamental factors on the pricing of shares of Russian companies. The study was conducted using its own 5-factor basic pricing model based on a sample of the most complete number of issues of shares of Russian issuers and a long time horizon, from 1997 to 2017. The market portfolio was the widest for a set of issuers. We consider the factor model as a kind of universal indicator of the efficiency of the stock market performance of its functions. The article confirms the significance of factors of a broad market portfolio, size, liquidity and, in part, momentum (inertia). However, starting from 2011, the significance of factors began to decrease as the qualitative characteristics of the stock market deteriorated due to the outflow of foreign portfolio investment, combined with the low level of development of domestic institutional investors. Also identified is the cyclical nature of the actions of company size and liquidity factors. Their ability to generate additional income on shares rises mainly at the stage of the fall of the stock market. The results of the study suggest that as domestic institutional investors develop on the Russian stock market, factor investment strategies can be used as a tool to increase the return on investor portfolios.

Download Full-text

The relation between R&D accounting treatment and the risk of the firm: Evidence from the Italian market

FINANCIAL REPORTING ◽

10.3280/fr2014-001002 ◽

2014 ◽

pp. 33-54 ◽

Cited By ~ 1

Author(s):

Riccardo Cimini ◽

Alessandro Gaetano ◽

Alessandra Pagani

Keyword(s):

Earnings Management ◽

Fiscal Year ◽

Investment Strategies ◽

Capital Assets ◽

Italian Market ◽

D Investment ◽

Firm Evidence

In this paper, we investigate the relation between the different accounting treatments of R&D expenditures and the risk of the entity in order to identify under which treatment insiders are more likely to carry out earnings management. By analysing the R&D investment strategies of a sample of 137 listed Italian entities that complied with the requirements of IAS 38 during fiscal year 2009, following Lantz and Sahut (2005), we calculate several indexes that show the preferences of insiders to account R&D expenditures as costs or capital assets, and we study the relation of such preferences with the risk of the entity, which we measure with the unlevered beta. We hypothesize that the entities, which considered the R&D investments as costs, are the riskiest ones due to the higher probability that insiders carried out earnings management. Our results confirm such hypothesis. This paper could have implications for academics and standard setters that could learn that behind accounting discretion, insiders could opportunistically behave against outsiders.

Download Full-text