Forecast Density Combinations of Dynamic Models and Data Driven Portfolio Strategies (original) (raw)
Related papers
Optimal Asset Allocation with Multivariate Bayesian Dynamic Linear Models
SSRN Electronic Journal, 2018
We introduce a simulation-free method to model and forecast multiple asset returns and employ it to investigate the optimal ensemble of features to include when jointly predicting monthly stock and bond excess returns. Our approach builds on the Bayesian Dynamic Linear Models of West and Harrison (1997), and it can objectively determine, through a fully automated procedure, both the optimal set of regressors to include in the predictive system and the degree to which the model coefficients, volatilities, and covariances should vary over time. When applied to a portfolio of five stock and bond returns, we find that our method leads to large forecast gains, both in statistical and economic terms. In particular, we find that relative to a standard no-predictability benchmark, the optimal combination of predictors, stochastic volatility, and time-varying covariances increases the annualized certainty equivalent returns of a leverage-constrained power utility investor by more than 500 basis points.
Computational Statistics & Data Analysis, 2014
We propose a Bayesian non-parametric approach for modeling the distribution of multiple returns. In particular, we use an asymmetric dynamic conditional correlation (ADCC) model to estimate the time-varying correlations of financial returns where the individual volatilities are driven by GJR-GARCH models. The ADCC-GJR-GARCH model takes into consideration the asymmetries in individual assets' volatilities, as well as in the correlations. The errors are modeled using a Dirichlet location-scale mixture of multivariate Gaussian distributions allowing for a great flexibility in the return distribution in terms of skewness and kurtosis. Model estimation and prediction are developed using MCMC methods based on slice sampling techniques. We carry out a simulation study to illustrate the flexibility of the proposed approach. We find that the proposed DPM model is able to adapt to several frequently used distribution models and also accurately estimates the posterior distribution of the volatilities of the returns, without assuming any underlying distribution. Finally, we present a financial application using Apple and NASDAQ Industrial index data to solve a portfolio allocation problem. We find that imposing a restrictive parametric distribution can result into underestimation of the portfolio variance, whereas DPM model is able to overcome this problem.
A Dynamic Conditional Approach to Portfolio Weights Forecasting
SSRN Electronic Journal
We build the time series of optimal realized portfolio weights from highfrequency data and we suggest a novel Dynamic Conditional Weights (DCW) model for their dynamics. DCW is benchmarked against popular model-based and model-free specifications in terms of weights forecasts and portfolio allocations. Next to portfolio variance, certainty equivalent and turnover, we introduce the break-even transaction costs as an additional measure that identifies the range of transaction costs for which one allocation is preferred to another. By comparing minimum-variance portfolios built on the components of the Dow Jones 30 Index, the proposed DCW overall attains the best allocations with respect to the measures considered, for any degree of risk-aversion, transaction costs and exposure.
Bayesian Non-Parametric Portfolio Decisions with Financial Time Series
A Bayesian non-parametric approach for efficient risk management is proposed. A dynamic model is considered where optimal portfolio weights and hedging ratios are adjusted at each period. The covariance matrix of the returns is described using an asymmetric MGARCH model. Restrictive parametric assumptions for the errors are avoided by relying on Bayesian nonparametric methods, which allow for a better evaluation of the uncertainty in financial decisions. Illustrative risk management problems using real data are solved. Significant differences in posterior distributions of the optimal weights and ratios are obtained arising from different assumptions for the errors in the time series model.
Bayesian Filtering for Multi-period Mean-Variance Portfolio Selection
2020
For a long investment time horizon, it is preferable to rebalance the portfolio weights at intermediate times. This necessitates a multi-period market model in which portfolio optimization is usually done through dynamic programming. However, this assumes a known distribution for the parameters of the financial time series. We consider the situation where this distribution is unknown and needs to be estimated from the data that is arriving dynamically. We applied Bayesian filtering through dynamic linear models to sequentially update the parameters. We considered uncertain investment lifetime to make the model more adaptive to the market conditions. These updated parameters are put into the dynamic mean-variance problem to arrive at optimal efficient portfolios. Extensive simulations are conducted to study the effect of varying underlying parameters and investment horizon on the performance of the method. An implementation of this model to the S&P500 illustrates that the Bayesian up...
A Bayesian Dynamic Compositional Model for Large Density Combinations in Finance
SSRN Electronic Journal
A Bayesian dynamic compositional model is introduced that can deal with combining a large set of predictive densities. It extends the mixture of experts and the smoothly mixing regression models by allowing for combination weight dependence across models and time. A compositional model with Logistic-normal noise is specified for the latent weight dynamics and the class-preserving property of the logistic-normal is used to reduce the dimension of the latent space and to build a compositional factor model. The projection used in the dimensionality reduction is based on a dynamic clustering process which partitions the large set of predictive densities into a smaller number of subsets. We exploit the state space form of the model to provide an efficient inference procedure based on Particle MCMC. The approach is applied to track the Standard & Poor 500 index combining 3712 predictive densities, based on 1856 US individual stocks, clustered in relatively small number of model sets. For the period 2007-2009, which included the financial crisis, substantial predictive gains are obtained, in particular, in the tails using Value-at-Risk. Similar predictive gains are obtained for the US Treasury Bill yield using a large set of macroeconomic variables. Evidence obtained on model set incompleteness and dynamic patterns in the financial clusters provide valuable signals for improved modelling and more effective economic and financial decisions.
Predictive Gains from Forecast Combinations Using Time Varying Model Weights
SSRN Electronic Journal, 2000
Several frequentist and Bayesian model averaging schemes, including a new one that simultaneously allows for parameter uncertainty, model uncertainty and time varying model weights, are compared in terms of forecast accuracy over a set of simulation experiments. Artificial data are generated, characterized by low predictability, structural instability, and fat tails, which is typical for many financial-economic time series. Sensitivity of results with respect to misspecification of the number of included predictors and the number of included models is explored. Given the set up of our experiments, time varying model weight schemes outperform other averaging schemes in terms of predictive gains both when the correlation among individual forecasts is low and the underlying data generating process is subject to structural locations shifts. In an empirical application using returns on the S&P 500 index, time varying model weights provide improved forecasts with substantial economic gains in an investment strategy including transaction costs.
Bayesian Dynamic Factor Models and Portfolio Allocation
Journal of Business & Economic Statistics, 2000
We discuss the development of dynamic factor models for multivariate financial time series, and the incorporation of stochastic volatility components for latent factor processes. Bayesian inference and computation is developed and explored in a study of the dynamic factor structure of daily spot exchange rates for a selection of international currencies. The models are direct generalisations of univariate stochastic volatility models, and represent specific varieties of models recently discussed in the growing multivariate stochastic volatility literature. We discuss model fitting based on retrospective data and sequential analysis for forward filtering and short-term forecasting. Analyses are compared with results from the much simpler method of dynamic variance matrix discounting that, for over a decade, has been a standard approach in applied financial econometrics. We study these models in analysis, forecasting and sequential portfolio allocation for a selected set of international exchange rate return time series. Our goals are to understand a range of modelling questions arising in using these factor models, and to explore empirical performance in portfolio construction relative to discount approaches. We report on our experiences and conclude with comments about the practical utility of structured factor models, and on future potential model extensions.
Forecast accuracy and economic gains from Bayesian model averaging using time-varying weights
Journal of Forecasting, 2010
Several Bayesian model combination schemes, including some novel approaches that simultaneously allow for parameter uncertainty, model uncertainty and robust time varying model weights, are compared in terms of forecast accuracy and economic gains using financial and macroeconomic time series. The results indicate that the proposed time varying model weight schemes outperform other combination schemes in terms of predictive and economic gains. In an empirical application using returns on the S&P 500 index, time varying model weights provide improved forecasts with substantial economic gains in an investment strategy including transaction costs. Another empirical example refers to forecasting US economic growth over the business cycle. It suggests that time varying combination schemes may be very useful in business cycle analysis and forecasting, as these may provide an early indicator for recessions.