How Often Should You Take Tactical Asset Allocation Decisions?
Byeong-Je An, et al.
March 5, 2015
About once a quarter. We compute optimal tactical asset allocation (TAA) policies over equities and bonds when both asset returns are predictable. By varying how often the weights are reset, we estimate the benefits and costs of different frequencies of TAA decisions. Tactical tilts taking advantage of predictable stock returns generate approximately twice as much value as those market-timing bond returns.
A Century of Generalized Momentum; From Flexible Asset Allocations (FAA) to Elastic Asset Allocation (EAA)
Wouter J. Keller and Adam Butler
December 30, 2014
This paper follows Keller (2012), which introduced the Flexible Asset Allocation (FAA) concept. FAA is based on a weighted ranking score of historical asset returns (R), volatilities (V), and correlations to an equal weighted index (C). We call this “generalized momentum” since we assume persistence in the short-term, not only for R, but also for V and C. Portfolios were formed monthly from a specified quantile of assets with the highest combined score.
In this paper we generalize FAA, starting from a tactical version of Modern Portfolio Theory (MPT) proposed in Keller (2013). Instead of choosing assets in the portfolio by a weighted ordinal rank on R, V, and C as in FAA, our new methodology – called Elastic Asset Allocation (EAA) – uses a geometrical weighted average of the historical returns, volatilities and correlations, using elasticities as weights. In order to avoid datasnooping (or curvefitting), we optimize the EAA model exclusively during a 50-year in-sample period (IS) from 1914 and apply these optimal IS parameters to test the model during an out-of-sample (OS) period from 1964-2014. The EAA model demonstrates impressive risk-adjusted and absolute OS performance over an equal weighted index for a variety of global asset universes.
Optimal Asset Allocation Across Investment Horizons
Ronald W. Best, et al.
January 30, 2015
We investigate the optimal portfolio mix of bonds and stocks across investment horizons. Sharpe ratios are computed using simulated returns for portfolios ranging from 100% bonds to 100% stocks where the portfolio mix is varied in increments of five percentage points. Holding periods from one to 25 years are examined under alternate return correlation assumptions. The optimal mix of bonds and stocks is identified as the portfolio with the highest Sharpe ratio for each holding period. The results show that optimal asset allocation varies across investment horizons and depends on whether security returns are independent or autocorrelated. If returns are independent over time, the weight of bonds increases as the investment horizon lengthens. If returns are autocorrelated over time, the weight of stocks increases as the investment horizon grows longer.
Credit Risk Premium: Its Existence and Implications for Asset Allocation
Attakrit Asvanunt and Scott A. Richardson
March 16, 2015
Using data from both cash bond markets (1927-2014) and synthetic CDS markets (2004-2014), we document evidence of a sizable credit risk premium. This premium is related to, but distinct from, the well-known equity risk premium and term premium. We further document variation in the size of the credit risk premium across different macroeconomic regimes: the credit risk premium is larger during periods of economic growth. Our empirical analyses support a strategic allocation to corporate credit and the possibility of a tactical allocation by exploiting forecasts of expected growth and aggregate default rates.
Momentum, Markowitz, and Smart Beta: A Tactical, Analytical and Practical Look at Modern Portfolio Theory
Wouter J. Keller
June 13, 2014
In this paper we will try to improve on the Modern Portfolio Theory (MPT) as developed by Markowitz (1952). As a first step, we combine the MPT model with generalized momentum (see Keller 2012) in order to arrive at a “tactical” MPT. In our second step, we will use the single index model (Elton, 1976) to arrive at an analytical solution for a long-only maximum Sharpe allocation. We will call this the MAA model, for Modern Asset Allocation. In our third step, we use shrinkage estimators in our formula for asset returns, volatilities and correlations to arrive at practical allocations. In addition, as a special cases, we arrive at EW (Equal Weight), Minimum Variance (MV), Maximum Diversification (MD) and (naïve) Risk Parity (RP) submodels of MAA. These EW, MV, MD and RP models are sometimes called “smart-beta” models. We illustrate all these different models on three universes consisting of respectively 10 and 35 global ETFs, and 104 US stocks/bonds, with daily data from Jan. 1998-Dec. 2013 (16 years), monthly rebalanced. We show that all these models beat the simple EW model con-sistently on various return /risk criteria, with the general MAA model (with return momentum) also beats nearly all of the “smart beta” models.