Ivilina Popova

Jackknife estimators for reducing bias in asset allocation

We use jackknife-based estimators to reduce bias when estimating the optimal value of a stochastic program. Our discussion focuses on an asset allocation model with a power utility function. As we will describe, estimating the optimal value of such a problem plays a key role in establishing the quality of a candidate solution, and reducing bias improves our ability to do so efficiently. We develop a jackknife estimator that is adaptive in that it does not assume the order of the bias is known a priori.

Higher-Order Moments of Fundamentals: A Literature Review

This literature review outlines the major progress in the research of the fundamental higher-order moments. We survey the existence, the formation, and the financial market and macroeconomics implications for the moments. Research shows that the time-varying volatility and the non-Gaussian shocks widely exist in all measures of fundamentals at both micro and macro levels. Additionally, the granular network among firms is the origin of the fundamental higher-order moments. Empirical evidence shows that the moments have strong predictive power on asset prices and macroeconomic variables. We also highlight crucial areas where more research is still needed to better understand the moments.

Estimation of Performance and Execution Time Effect on High Frequency Statistical Arbitrage Strategies

This article is designed to help quantify one of the “slippages” that are often recognized in quant strategies. The idea is that whenever the actual executed prices are away (both time and size) from the model prices, the realized returns suffer. The slippage for a particular statistical arbitrage strategy is quantified. It is shown that a portion of the loss is due to using different prices for estimating the parameters of the strategy. The main source of the loss is the use of intraday in place of market on close prices. Five years of intraday transaction data from the NYSE TAQ database are used. Analysis shows that on average the daily loss due to intraday prices accounts for 0.03% of the initial capital. For the period 2003 through 2006, the accumulated loss is approximately 30%. The described approach can be of use to new quantitative analysts who create and backtest trading strategies. It could also be used during the due diligence process of a fund that is interested in investing in a statistical arbitrage strategy. This article recommends requiring that a backtest be done by using intraday and market on close prices in order to identify the size of such loss.

Simulating cointegrated time series

When one models dependence solely via correlations, portfolio allocation models can perform poorly. This motivates considering dependence measures other than correlation. Cointegration is one such measure that captures long-term dependence. In this paper we present a new method to simulate cointegrated sample paths using the vector auto-regressive-to-anything (VARTA) algorithm. Our approach relies on new properties of cointegrated time series of financial asset prices and allows for marginal distributions from the Johnson system. The method is illustrated on two data sets, one real and one artificial.