Unleveraged Portfolios and Pure Allocation Return

In asset management, the portfolio leverage affects performance, and can be subject to constraints and operational limitations. Due to the possible leverage aversion of the investors, the comparison between portfolio performances can be incomplete or misleading. We propose a procedure to unleverage the mean-variance efficient portfolios to satisfy a leverage requirement. We obtain a class of unleveraged portfolios that are homogeneous in terms of leverage, so therefore properly comparable. The proposed unleverage procedure permits isolating the pure allocation return, i.e., the return component, due to the qualitative choice of portfolio allocation, from the return component due to the portfolio leverage. Theoretical analysis and empirical evidence on actual data show that efficient mean-variance portfolios, once unleveraged, uncover mean-variance dominance relations hidden by the leverage contribution to portfolio return. Our approach may be useful to practitioners proposing to take long positions on “short assets” (e.g. inverse ETF), thereby considering short positions as active investment choices, in contrast with the usual interpretation where are used to overweight long positions.

Deep Reinforcement Learning for Stock Recommendation

Recommending stocks is very important for investment companies and investors. However, without enough analysts, no stock selection strategy can capture the dynamics of all S&P 500 stocks. Nevertheless, most existing recommending strategies are based on predictive models to buy and hold stocks with high return potential. But these strategies fail to recommend stocks from different industrial sectors to reduce risks. In this article, we propose a novel solution that recommends a stock portfolio with reinforcement learning from the S&P 500 index. Our basic idea is to construct a stock relation graph (RG) which provide rich relations among stocks and industrial sectors, to generate diversified recommendation result. To this end, we design a new method to explore high-quality stocks from the constructed relation graph with reinforcement learning. Specifically, the reinforcement learning agent jumps from each industrial sector to select stock based on the feedback signals from the market. Finally, we apply portfolio allocation methods (i.e., mean-variance and minimum-variance) to test the validity of the recommendation. The empirical results show that the performance of portfolio allocation based on the selected stocks is better than the long-term strategy on the S&P 500 Index in terms of cumulative returns.

Analyst Dividend Forecasts and Their Usefulness to Investors

In contrast to the disappearing dividends view prevalent in the literature, we document extensive dividend payments by firms and significant variability within firms and across 16 countries during 2000-2013. We predict that within-firm variability in dividends increases investor demand for forward-looking dividend information, and analysts respond by producing informative dividend forecasts. We find that analyst dividend forecasts are available for most dividend-paying firms and are more prevalent for firms with higher variability of dividends. Analyst dividend forecasts are more accurate than alternative proxies based on extrapolations of past dividends. Finally, dividend forecasts (i) are incrementally useful to investors beyond information in other fundamentals such as earnings and cash flow forecasts, (ii) help investors interpret earnings quality, and (iii) are associated with investors' portfolio allocation decisions.

High-Frequency Cryptocurrency Trading Strategy using Tweet Sentiment Analysis

Sentiments are extracted from tweets with the hashtag of cryptocurrencies to predict the price and sentiment prediction model generates the parameters for optimization procedure to make decision and re-allocate the portfolio in the further step. Moreover, after the process of prediction, the evaluation, which is conducted with RMSE, MAE and R2, select the KNN and CART model for the prediction of Bitcoin and Ethereum respectively. During the process of portfolio optimization, this project is trying to use predictive prescription to robust the uncertainty and meanwhile take full advantages of auxiliary data such as sentiments. For the outcome of optimization, the portfolio allocation and returns fluctuate acutely as the illustration of figure.

Emerging market capital flows: the role of fund manager portfolio allocation

Empleamos datos desagregados de las tenencias de los fondos de inversión globales para distinguir entre las dos razones por las que pueden cambiar las tenencias de bonos de economías de mercado emergentes (EMEs) por parte de los fondos: (i) el monto invertido en el fondo puede cambiar y (ii) el administrador del fondo puede modificar la asignación del portafolio. Encontramos que la respuesta de los fondos a las condiciones macroeconómicas globales, "push factors", se explica por las decisiones de los inversionistas del fondo. Por otro lado, la respuesta de los fondos a las condiciones macroeconómicas locales, "pull factors", se explica por las reasignaciones en las tenencias por cuenta de los administradores de los fondos. Adicionalmente, identificamos otros factores instituciones que impactan las decisiones de reasignación: cambios en el apalancamiento, su índice de referencia, y su apetito de riesgo (los fondos reasignan recursos hacia EMEs más seguros ante incrementos en factores globales de riesgo).