Journal of Empirical Finance ( IF 3.025 ) Pub Date : 2021-05-10 , DOI: 10.1016/j.jempfin.2021.05.001 Azam Shamsi Zamenjani
This paper investigates predictability of the full density of market returns. The proposed model is a Bayesian nonparametric mixture model where the mixture weights are functions of predictors, allowing us to study predictability of the unknown and time-varying density of market returns, in contrast to the extant literature which essentially focuses on point forecasts of the predictive mean which contains no description of the associated uncertainty. We compare statistical and economic performance of the proposed model with a set of competing models. Despite little or no improvement in point forecasts, certain variables display significant out-of-sample predictive ability for the stock return density and increase economic value for investors when employed in portfolio decisions.
中文翻译:
财务变量是否有助于预测市场组合的条件分布?
本文研究了市场收益全密度的可预测性。所提出的模型是贝叶斯非参数混合模型,其中混合权重是预测变量的函数,从而使我们能够研究市场收益的未知和随时间变化的密度的可预测性,而现有文献主要侧重于预测变量的点预测。平均值,不包含相关不确定性的描述。我们将提议的模型的统计和经济绩效与一组竞争模型进行比较。尽管点预测几乎没有改善或没有改善,但某些变量显示出对股票收益密度的明显超出样本的预测能力,并在投资组合决策中使用时为投资者增加了经济价值。