Multiple Asset Classes Research Articles

New Evidence on Optimal Asset Allocation

AbstractBrocato and Steed (1998) showed that portfolio rebalancing based on NBER business cycle turning points substantially improves in‐sample Markowitz efficiency. In a similar vein, we investigate potential improvements from rebalancing based on turning points in the monetary cycle. We find that the monetary cycle has greater influence than the business cycle on the variance/covariance structure of multiple asset classes. Furthermore, we find substantial improvements in in‐sample efficiency beyond a buy‐and‐hold strategy and the business‐cycle approach. Importantly, our indicator of monetary cycle turning points has a practical advantage over NBER business cycle turning points, in that it relies only on ex ante information. In out‐of‐sample tests, we continue to find superior portfolio performance after transactions costs using the monetary cycle to time portfolio rebalancing.

Forecasting portfolio risk in normal and stressed markets

Forecasting portfolio risk in normal and stressed markets

Asset Allocation Dynamics and Pension Fund Performance

Using a data set on more than 300 U.K. pension funds' asset holdings, this article provides a systematic investigation of the performance of managed portfolios across multiple asset classes. We find evidence of slow mean reversion in the funds' portfolio weights toward a common, time-varying strategic asset allocation. We also find surprisingly little cross-sectional variation in the average ex post returns arising from the strategic-asset-allocation, market-timing, and security-selection decisions of the fund managers. Strategic asset allocation accounts for most of the time-series variation in portfolio returns, while market timing and asset selection appear to have been far less important. Copyright 1999 by University of Chicago Press.