Morningstar Adds to Direct Platform

Morningstar Inc. has added a new asset allocation and forecasting functionality to Morningstar Direct, a Web-based global investment analysis platform for institutional investors.

The addition will allow investors to create asset allocation strategies that account for “fat tails” and are optimized to mitigate downside risk. The new tools were developed based research from Morningstar and its wholly owned subsidiary, Ibbotson Associates, Inc.  

As the standard for creating efficient asset allocation strategies for more than fifty years, Morningstar explained that mean-variance optimization (MVO) has become synonymous with Modern Portfolio Theory. Traditional MVO, however, cannot take into account “fat-tailed” or extreme asset class return distributions, which better match real-world historical asset class returns. This framework is also limited by its ability to only optimize asset mixes for one risk metric (standard deviation) and one reward metric (arithmetic return).  

The new capabilities allow users to choose from a number of return distribution assumptions to model asset class behavior, including traditional bell-curve shaped return distributions as well as “fat-tailed” and skewed distributions. Users can then use scenario-based optimization to create optimal asset allocation strategies. They can elect to create strategies that will produce the highest expected return either for a given level of volatility or for one of several downside risk measures. The new asset allocation functionality is included with all Morningstar Direct subscriptions. 

“As the markets have shown—and reminded us most harshly again in 2008—real-life finance is often more complex than the traditional mathematical models used for portfolio optimization and forecasting,” said Xiaohua Xia, president of institutional software for Morningstar. “Our new asset allocation functionality in Morningstar Direct offers institutional investors a flexible tool with multiple options to enhance and refine traditional mean variance optimization or take advantage of some of the most cutting-edge modeling.” 

More information is available here