A UK company initially founded in 1990, EM Applications (EMA) now develops software tools for investment professionals to assist with risk analysis, portfolio construction and alpha generation. The various applications all rely on EMA’s use of  “Expectations Maximisation” factor analysis to model the world’s equity and bond markets.

The significant advantage of EMA’s purely statistical approach to modeling markets is that it is not dependent on the accuracy or availability of fundamental data and so responds naturally to locate new influences on market risk as they occur.  Whereas fundamental models are often restricted to monthly data sampling and have to adjust accounting data to take account of national differences. EMA’s statistical approach only requires performance data as input, allowing for easy comparability between markets and for frequent sampling. A further advantage of EMA’s Factor Analytical approach by comparison with the Principle Components technique often used for purely statistical models is that it creates factors that are more stable over time.

Current products of EM Applications include:

Excerpt – the low cost equity risk analysis product that can generate live tracking errors, country, sector and style risk exposures and marginal contributions to risk. Excerpt also incorporates an easy to use optimizer which can be used to build efficient portfolios or simply for trading idea generation.

Estema – a powerful factor modeling program that allows you to build your own risk models using EM factor analysis. Such models can be built with just statistical factors or with any mix of statistical and pre-specified factors.

Key staff of EM Applications are:

Professor Al Stroyny – Chairman - has worked with factor analysis algorithms since 1980.  He has developed robust methods of maximum likelihood factor analysis and applied such techniques to large data sets of stock returns.  Dr Stronyny has a Ph.D. from the University of Wisconsin on “Heteroskedasticity and Estimation of Sytematic Risk” and has taught finance at Marquette University and the University of Wisconsin.  He has also worked at Yamaichi, Fortis and the Bank of New York.

Dr Tim Wilding – (joined 1994) Head of Research and Development, with a Ph.D. from the Department of Physics at the University of Cambridge. He has spent the last 10 years applying skills gained from his doctorate working for investment risk analysis firms. At EM Applications, he has developed general-purpose estimation routines to fit any type of model.  He has also developed propriety extensions to factor models that allow them to be used with different asset classes, and he has also built tools for portfolio optimisation. These tools include advanced techniques for quadratic programming and genetic algorithms.

Peter Ainsworth – Managing Director - has 20 years experience in the financial markets including portfolio trading, broking and marketing. He was a Senior Vice President with Merrill Lynch in the 1980’s, becoming the in-house expert in the application of Merrill Lynch’s Quantitative Product and Barra’s US Equity product to the needs of London based institutions. He joined Drexel Burnham Lambert in 1989 as Head of Quantitative Marketing where he worked with the “APT” (Advanced Portfolio Technologies) risk system. After Drexel went into liquidation in 1990 he set up APT UK Ltd to distribute the APT system in Europe and also FactSet (UK) Ltd, the forerunner to EMA. After being introduced to Al Stroyny in 1994 he terminated the relationship with APT and converted the business of FactSet (UK) Ltd to the distribution of applications based on Stroyny’s work.

EM has offices in the US and UK and its main product is Excerpt.

You can find out more about EMA by visiting their web site at www.emapplications.com

quant shop pty ltd