TPLVM: Portfolio Construction by Student's t-process Latent Variable Model

  title={TPLVM: Portfolio Construction by Student's t-process Latent Variable Model},
  author={Yusuke Uchiyama and Kei Nakagawa},
Optimal asset allocation is a key topic in modern finance theory. To realize the optimal asset allocation on investor's risk aversion, various portfolio construction methods have been proposed. Recently, the applications of machine learning are rapidly growing in the area of finance. In this article, we propose the Student's $t$-process latent variable model (TPLVM) to describe non-Gaussian fluctuations of financial timeseries by lower dimensional latent variables. Subsequently, we apply the… 
GO-GJRSK Model with Application to Higher Order Risk-Based Portfolio
There are three distinguishing features in the financial time series, such as stock prices, are as follows: (1) Non-normality, (2) serial correlation, and (3) leverage effect. All three points need
How Do We Predict Stock Returns in the Cross-Section with Machine Learning?
Investigation of how the difference in problem settings such as problem definition and data preprocessing affects the performance of stock return prediction shows that the performance varies depending on problem settings regardless of the prediction models.
Entropy Based Student’s t-Process Dynamical Model
An entropy based Student’s t-process Dynamical model (ETPDM) is proposed as a volatility fluctuation model combined with both nonlinear dynamics and non-Gaussian noise to test the performance of the ETPDM.


Applications of Gaussian Process Latent Variable Models in Finance
This work proposes a novel covariance estimator based on the Gaussian Process Latent Variable Model (GP-LVM), which can be considered as a non-linear extension of standard factor models with readily interpretable parameters reminiscent of market betas.
Risk-Based Portfolios with Large Dynamic Covariance Matrices
In the field of portfolio management, practitioners are focusing increasingly on risk-based portfolios rather than on mean-variance portfolios. Risk-based portfolios are constructed based solely on
Portfolio Choices with Orthogonal Bandit Learning
This paper presents a bandit algorithm for conducting online portfolio choices by effectually exploiting correlations among multiple arms and derives the optimal portfolio strategy that represents the combination of passive and active investments according to a risk-adjusted reward function.
Toward Maximum Diversification
Along with the ongoing effort to build market cap–independent portfolios, the authors explore the properties of diversification as a driver of portfolio construction. They introduce a measure of the
Deep Recurrent Factor Model: Interpretable Non-Linear and Time-Varying Multi-Factor Model
The linear multi-factor model is extended to be non-linear and time-varying with LSTM, and the recurrent model has better predictive capability than the traditional linear model and fully-connected deep learning methods.
Nonlinear Shrinkage of the Covariance Matrix for Portfolio Selection: Markowitz Meets Goldilocks
Markowitz (1952) portfolio selection requires an estimator of the covariance matrix of returns. To address this problem, we promote a nonlinear shrinkage estimator that is more flexible than previous
Large Dynamic Covariance Matrices
Second moments of asset returns are important for risk management and portfolio selection. The problem of estimating second moments can be approached from two angles: time series and the
Deep Factor Model - Explaining Deep Learning Decisions for Forecasting Stock Returns with Layer-Wise Relevance Propagation
This work proposes to represent a return model and risk model in a unified manner with deep learning, which is a representative model that can express a nonlinear relationship and shows that the deep factor model has better predictive capability than the traditional linear model or other machine learning methods.
Gaussian Process Volatility Model
GP-Vol is introduced, a novel non-parametric model for time-changing variances based on Gaussian Processes that can capture highly flexible functional relationships for the variances and is much faster than current offline inference procedures and avoids overriding problems by following a fully Bayesian approach.