Extended Viewer

Imre Balint

Research and Insights

Articles by Imre Balint

    Using Systematic Equity Strategies: Managing Active Portfolios in the Global Equity Universe

    Research Report | Apr 18, 2016 | Dimitris Melas, Imre Balint

    Both quantitative managers and fundamental stockpickers need to understand their risk exposures, build efficient portfolios and differentiate themselves from their competitors. Factor models identify country, industry and style factors, which help forecast and explain portfolio risk. A subset of style factors, Systematic Equity Strategy (SES) factors — such as value, momentum and quality — has also earned positive long-term returns historically. In this paper, we review the role of SES...

    Research Insight - Riding on Momentum

    Research Report | Dec 15, 2015 | Abhishek Gupta, Dimitris Melas, Imre Balint, Jain Vipul

    Momentum, the tendency of past winners to continue to do well in the near future, is used widely in risk models, quantitative strategies and, more recently, as the basis for factor indexes aiming to replicate the performance of this pervasive factor. But the academic definition of momentum is extremely difficult to implement because it tends to lead to high volatility exposure and excessive portfolio turnover. Our approach involves selecting securities based on risk-adjusted performance and...

    Should You Care About Active Share

    Research Report | Jul 14, 2015 | Altaf Kassam, Imre Balint

    A Portfolio Construction Study”Active Share” has been widely credited as a predictor of manager skill. Initial academic research has shown that active managers with high Active Share and low Tracking Error enjoyed persistent outperformance. Conversely, managers with low Active Share scores and low Tracking Error were labelled “closet indexers” and have recently become the subject of regulatory scrutiny.  But how reliable is Active Share as a single metric? ...

    Model Insight - The Barra Emerging Markets Model Empirical Notes - February 2014

    Research Report | Feb 25, 2014 | Andrei Morozov, Mehmet Bayraktar, Imre Balint, Laszlo Borda, Paul Ward

    This Model Insight provides empirical results for the new Barra Emerging Markets Model, including detailed information on the structure, the performance, and the explanatory power of the factors. Furthermore, these notes also include backtesting results and a thorough side-by-side comparison of the forecasting accuracy of the new Emerging Markets Model and the Global Equity Model (GEM3).

    Model Insight - Barra North America Stochastic Factor Model (NAMS1) Research Notes - April 2013

    Research Report | Apr 12, 2013 | Andrei Morozov, Imre Balint, Paul Ward

    The Barra North America Stochastic Model (NAMS1) applies the stochastic methodology framework to the US and Canada equity markets. These detailed Research Notes discuss an overview of the model specifications, offer insights into the model behavior and applications, and provide the results of extensive backtests from 1995-2012.

    Model Insight - Barra North America Stochastic Factor Model (NAMS1) Highlights - February 2013

    Research Report | Feb 24, 2013 | Andrei Morozov, Imre Balint, Paul Ward

    The Barra North America Stochastic Factor Model (NAMS1) is the second in a family of statistical factor models developed by MSCI, following the launch of the Barra Europe Stochastic Factor Model (EURS1) in September 2012. This overview document succinctly describes the NAMS1 specifications and provides the results of extensive backtests from 1995-2012.

    What Makes a Good Statistical Model?

    Research Report | Dec 7, 2012 | Imre Balint, Rachael Smith

    In this paper, we investigate whether the new Barra Europe Stochastic Factor Model (EURS1) performs differently in portfolio construction when compared to other statistical models commonly used by investors.  In order to explore this, we built a typical principal component analysis (PCA) model and used both the PCA and EURS1 models to track two popular benchmarks using the Barra Open Optimizer.  In all cases, we found EURS1 to have the lowest tracking error, and the EURS1 slow model...