Quantitative Finance

Portfolio traders strive to identify dynamic portfolio allocation schemes so that their total budgets are efficiently allocated through the investment horizon. This study proposes a novel portfolio trading strategy in which an intelligent agent is trained to identify an optimal trading action by using deep Q-learning. We formulate a Markov decision process model for the portfolio trading process, and the model adopts a discrete combinatorial action space, determining the trading direction at prespecified trading size for each asset, to ensure practical applicability. Our novel portfolio trading strategy takes advantage of three features to outperform in real-world trading. First, a mapping function is devised to handle and transform an initially found but infeasible action into a feasible action closest to the originally proposed ideal action. Second, by overcoming the dimensionality problem, this study establishes models of agent and Q-network for deriving a multi-asset trading strategy in the predefined action space. Last, this study introduces a technique that has the advantage of deriving a well-fitted multi-asset trading strategy by designing an agent to simulate all feasible actions in each state. To validate our approach, we conduct backtests for two representative portfolios and demonstrate superior results over the benchmark strategies.

In the past decade many researchers have proposed new optimal portfolio selection strategies to show that sophisticated diversification can outperform the naïve 1/N strategy in out-of-sample benchmarks. Providing an updated review of these models since DeMiguel et al. (2009b), I test sixteen strategies across six empirical datasets to see if indeed progress has been made. However, I find that none of the recently suggested strategies consistently outperforms the 1/N or minimum-variance approach in terms of Sharpe ratio, certainty-equivalent return or turnover. This suggests that simple diversification rules are not in fact inefficient, and gains promised by optimal portfolio choice remain unattainable out-of-sample due to large estimation errors in expected returns. Therefore, further research effort should be devoted to both improving estimation of expected returns, and possibly exploring diversification rules that do not require the estimation of expected returns directly, but also use other available information about the stock characteristics.

Global fixed income returns span across multiple maturities and economies, that is, they naturally reside on multi-dimensional data structures referred to as tensors. In contrast to standard "flat-view" multivariate models that are agnostic to data structure and only describe linear pairwise relationships, we introduce a tensor-valued approach to model the global risks shared by multiple interest rate curves. In this way, the estimated risk factors can be analytically decomposed into maturity-domain and country-domain constituents, which allows the investor to devise rigorous and tractable global portfolio management and hedging strategies tailored to each risk domain. An empirical analysis confirms the existence of global risk factors shared by eight developed economies, and demonstrates their ability to compactly describe the global macroeconomic environment.

The optimization of the variance supplemented by a budget constraint and an asymmetric ℓ 1 regularizer is carried out analytically by the replica method borrowed from the theory of disordered systems. The asymmetric regularizer allows us to penalize short and long positions differently, so the present treatment includes the no-short-constrained portfolio optimization problem as a special case. Results are presented for the out-of-sample and the in-sample estimator of the regularized variance, the relative estimation error, the density of the assets eliminated from the portfolio by the regularizer, and the distribution of the optimal portfolio weights. We have studied the dependence of these quantities on the ratio r of the portfolio's dimension N to the sample size T , and on the strength of the regularizer. We have checked the analytic results by numerical simulations, and found general agreement. Regularization extends the interval where the optimization can be carried out, and suppresses the large sample fluctuations, but the performance of ℓ 1 regularization is rather disappointing: if the sample size is large relative to the dimension, i.e. r is small, the regularizer does not play any role, while for r 's where the regularizer starts to be felt the estimation error is already so large as to make the whole optimization exercise pointless. We find that the ℓ 1 regularization can eliminate at most half the assets from the portfolio, corresponding to this there is a critical ratio r=2 beyond which the ℓ 1 regularized variance cannot be optimized: the regularized variance becomes constant over the simplex. These facts do not seem to have been noticed in the literature.

A large portfolio of independent returns is optimized under the variance risk measure with a ban on short positions. The no-short selling constraint acts as an asymmetric ??1 regularizer, setting some of the portfolio weights to zero and keeping the out of sample estimator for the variance bounded, avoiding the divergence present in the non-regularized case. However, the susceptibility, i.e. the sensitivity of the optimal portfolio weights to changes in the returns, diverges at a critical value r=2 . This means that a ban on short positions does not prevent the phase transition in the optimization problem, it merely shifts the critical point from its non-regularized value of r=1 to 2 . At r=2 the out of sample estimator for the portfolio variance stays finite and the estimated in-sample variance vanishes. We have performed numerical simulations to support the analytic results and found perfect agreement for N/T<2 . Numerical experiments on finite size samples of symmetrically distributed returns show that above this critical point the probability of finding solutions with zero in-sample variance increases rapidly with increasing N , becoming one in the large N limit. However, these are not legitimate solutions of the optimization problem, as they are infinitely sensitive to any change in the input parameters, in particular they will wildly fluctuate from sample to sample. We also calculate the distribution of the optimal weights over the random samples and show that the regularizer preferentially removes the assets with large variances, in accord with one's natural expectation.

This paper examines the optimal annuitization, investment and consumption strategies of a utility-maximizing retiree facing a stochastic time of death under a variety of institutional restrictions. We focus on the impact of aging on the optimal purchase of life annuities which form the basis of most Defined Benefit pension plans. Due to adverse selection, acquiring a lifetime payout annuity is an irreversible transaction that creates an incentive to delay. Under the institutional all-or-nothing arrangement where annuitization must take place at one distinct point in time (i.e. retirement), we derive the optimal age at which to annuitize and develop a metric to capture the loss from annuitizing prematurely. In contrast, under an open-market structure where individuals can annuitize any fraction of their wealth at anytime, we locate a general optimal annuity purchasing policy. In this case, we find that an individual will initially annuitize a lump sum and then buy annuities to keep wealth to one side of a separating ray in wealth-annuity space. We believe our paper is the first to integrate life annuity products into the portfolio choice literature while taking into account realistic institutional restrictions which are unique to the market for mortality-contingent claims.

Machine Learning algorithms and Neural Networks are widely applied to many different areas such as stock market prediction, face recognition and population analysis. This paper will introduce a strategy based on the classic Deep Reinforcement Learning algorithm, Deep Q-Network, for portfolio management in stock market. It is a type of deep neural network which is optimized by Q Learning. To make the DQN adapt to financial market, we first discretize the action space which is defined as the weight of portfolio in different assets so that portfolio management becomes a problem that Deep Q-Network can solve. Next, we combine the Convolutional Neural Network and dueling Q-net to enhance the recognition ability of the algorithm. Experimentally, we chose five lowrelevant American stocks to test the model. The result demonstrates that the DQN based strategy outperforms the ten other traditional strategies. The profit of DQN algorithm is 30% more than the profit of other strategies. Moreover, the Sharpe ratio associated with Max Drawdown demonstrates that the risk of policy made with DQN is the lowest.

In this paper, we empirically study models for pricing Italian sovereign bonds under a reduced form framework, by assuming different dynamics for the short-rate process. We analyze classical Cox-Ingersoll-Ross and Vasicek multi-factor models, with a focus on optimization algorithms applied in the calibration exercise. The Kalman filter algorithm together with a maximum likelihood estimation method are considered to fit the Italian term-structure over a 12-year horizon, including the global financial crisis and the euro area sovereign debt crisis. Analytic formulas for the gradient vector and the Hessian matrix of the likelihood function are provided.

Investors in Target Date Funds are automatically switched from high risk to low risk assets as their retirements approach. Such funds have become very popular, but our analysis brings into question the rationale for them. Based on both a model with parameters fitted to historical returns and on bootstrap resampling, we find that adaptive investment strategies significantly outperform typical Target Date Fund strategies. This suggests that the vast majority of Target Date Funds are serving investors poorly.

Artificial intelligence, or AI, enhancements are increasingly shaping our daily lives. Financial decision-making is no exception to this. We introduce the notion of AI Alter Egos, which are shadow robo-investors, and use a unique data set covering brokerage accounts for a large cross-section of investors over a sample from January 2003 to March 2012, which includes the 2008 financial crisis, to assess the benefits of robo-investing. We have detailed investor characteristics and records of all trades. Our data set consists of investors typically targeted for robo-advising. We explore robo-investing strategies commonly used in the industry, including some involving advanced machine learning methods. The man versus machine comparison allows us to shed light on potential benefits the emerging robo-advising industry may provide to certain segments of the population, such as low income and/or high risk averse investors.