Dynamic Asset Allocation

Target-Date Funds: A State-of-the-Art Review with Policy Applications to Chile’s Pension Reform ArXiv ID: 2504.17713 “View on arXiv” Authors: Fernando Suárez, José Manuel Peña, Omar Larré Abstract This review paper explores the evolution and implementation of target-date funds (TDFs), specifically focusing on their application within the context of Chile’s 2025 pension reform. The introduction of TDFs marks a significant shift in Chile’s pension system, which has traditionally relied on a multifund structure (essentially a target-risk funds system). We offer a comprehensive review of the theoretical foundations and practical considerations of TDFs, highlighting key challenges and opportunities for Chilean regulators and fund managers. Notably, we recommend that the glide path design should be dynamic, incorporating adjustments based on total accumulated wealth, with particular flexibility depending on each investor’s risk tolerance. Furthermore, we propose that the new benchmark for generational funds should feature a wide deviation band relative to the new benchmark portfolio, which could foster a market with more investment strategies and better competition among fund managers, encourage the inclusion of alternative assets, and foster greater diversification. Lastly, we highlight the need for future work to define a glide path model that incorporates the theoretical frameworks described, tailored to the unique parameters of the Chilean pension system. These recommendations aim to optimize the long-term retirement outcomes for Chilean workers under the new pension structure. ...

Multiscale Markowitz ArXiv ID: 2411.13792 “View on arXiv” Authors: Unknown Abstract Traditional Markowitz portfolio optimization constrains daily portfolio variance to a target value, optimising returns, Sharpe or variance within this constraint. However, this approach overlooks the relationship between variance at different time scales, typically described by $σ(Δt) \propto (Δt)^{“H”}$ where $H$ is the Hurst exponent, most of the time assumed to be (\frac{“1”}{“2”}). This paper introduces a multifrequency optimization framework that allows investors to specify target portfolio variance across a range of frequencies, characterized by a target Hurst exponent $H_{“target”}$, or optimize the portfolio at multiple time scales. By incorporating this scaling behavior, we enable a more nuanced and comprehensive risk management strategy that aligns with investor preferences at various time scales. This approach effectively manages portfolio risk across multiple frequencies and adapts to different market conditions, providing a robust tool for dynamic asset allocation. This overcomes some of the traditional limitations of Markowitz, when it comes to dealing with crashes, regime changes, volatility clustering or multifractality in markets. We illustrate this concept with a toy example and discuss the practical implementation for assets with varying scaling behaviors. ...

Improving Portfolio Optimization Results with Bandit Networks ArXiv ID: 2410.04217 “View on arXiv” Authors: Unknown Abstract In Reinforcement Learning (RL), multi-armed Bandit (MAB) problems have found applications across diverse domains such as recommender systems, healthcare, and finance. Traditional MAB algorithms typically assume stationary reward distributions, which limits their effectiveness in real-world scenarios characterized by non-stationary dynamics. This paper addresses this limitation by introducing and evaluating novel Bandit algorithms designed for non-stationary environments. First, we present the Adaptive Discounted Thompson Sampling (ADTS) algorithm, which enhances adaptability through relaxed discounting and sliding window mechanisms to better respond to changes in reward distributions. We then extend this approach to the Portfolio Optimization problem by introducing the Combinatorial Adaptive Discounted Thompson Sampling (CADTS) algorithm, which addresses computational challenges within Combinatorial Bandits and improves dynamic asset allocation. Additionally, we propose a novel architecture called Bandit Networks, which integrates the outputs of ADTS and CADTS, thereby mitigating computational limitations in stock selection. Through extensive experiments using real financial market data, we demonstrate the potential of these algorithms and architectures in adapting to dynamic environments and optimizing decision-making processes. For instance, the proposed bandit network instances present superior performance when compared to classic portfolio optimization approaches, such as capital asset pricing model, equal weights, risk parity, and Markovitz, with the best network presenting an out-of-sample Sharpe Ratio 20% higher than the best performing classical model. ...

Dynamic Asset Allocation with Asset-Specific Regime Forecasts ArXiv ID: 2406.09578 “View on arXiv” Authors: Unknown Abstract This article introduces a novel hybrid regime identification-forecasting framework designed to enhance multi-asset portfolio construction by integrating asset-specific regime forecasts. Unlike traditional approaches that focus on broad economic regimes affecting the entire asset universe, our framework leverages both unsupervised and supervised learning to generate tailored regime forecasts for individual assets. Initially, we use the statistical jump model, a robust unsupervised regime identification model, to derive regime labels for historical periods, classifying them into bullish or bearish states based on features extracted from an asset return series. Following this, a supervised gradient-boosted decision tree classifier is trained to predict these regimes using a combination of asset-specific return features and cross-asset macro-features. We apply this framework individually to each asset in our universe. Subsequently, return and risk forecasts which incorporate these regime predictions are input into Markowitz mean-variance optimization to determine optimal asset allocation weights. We demonstrate the efficacy of our approach through an empirical study on a multi-asset portfolio comprising twelve risky assets, including global equity, bond, real estate, and commodity indexes spanning from 1991 to 2023. The results consistently show outperformance across various portfolio models, including minimum-variance, mean-variance, and naive-diversified portfolios, highlighting the advantages of integrating asset-specific regime forecasts into dynamic asset allocation. ...

Causal Inference on Investment Constraints and Non-stationarity in Dynamic Portfolio Optimization through Reinforcement Learning ArXiv ID: 2311.04946 “View on arXiv” Authors: Unknown Abstract In this study, we have developed a dynamic asset allocation investment strategy using reinforcement learning techniques. To begin with, we have addressed the crucial issue of incorporating non-stationarity of financial time series data into reinforcement learning algorithms, which is a significant implementation in the application of reinforcement learning in investment strategies. Our findings highlight the significance of introducing certain variables such as regime change in the environment setting to enhance the prediction accuracy. Furthermore, the application of reinforcement learning in investment strategies provides a remarkable advantage of setting the optimization problem flexibly. This enables the integration of practical constraints faced by investors into the algorithm, resulting in efficient optimization. Our study has categorized the investment strategy formulation conditions into three main categories, including performance measurement indicators, portfolio management rules, and other constraints. We have evaluated the impact of incorporating these conditions into the environment and rewards in a reinforcement learning framework and examined how they influence investment behavior. ...

Optimal Investment with Stochastic Interest Rates and Ambiguity ArXiv ID: 2306.13343 “View on arXiv” Authors: Unknown Abstract This paper studies dynamic asset allocation with interest rate risk and several sources of ambiguity. The market consists of a risk-free asset, a zero-coupon bond (both determined by a Vasicek model), and a stock. There is ambiguity about the risk premia, the volatilities, and the correlation. The investor’s preferences display both risk aversion and ambiguity aversion. The optimal investment problem admits a closed-form solution. The solution shows that the ambiguity only affects the speculative motives of the investor, representing a hedge against the ambiguity, but not the hedging of interest rate risk. An implementation of the optimal investment strategy shows that ambiguity aversion helps to tame the highly leveraged portfolios neglecting ambiguity and leads to strategies that are more in line with popular investment advice. ...