{"title":"Gold risk premium estimation with machine learning methods","authors":"Juan D. Díaz , Erwin Hansen , Gabriel Cabrera","doi":"10.1016/j.jcomm.2022.100293","DOIUrl":null,"url":null,"abstract":"<div><p><span>This paper assesses the accuracy of several machine learning models’ predictions of the gold risk premium when using an extensive set of 186 predictors. We perform an out-of-sample evaluation and consider both statistical and portfolio metrics. Our results show that machine learning methods and forecast combinations have a limited ability to outperform the historical mean when predicting the gold risk premium. Slightly better results are obtained when predictors are used individually. More specifically, we find that several technical indicators (moving average and momentum series) have forecasting power during periods of expansion, while several business cycle variables and </span>geopolitical risk variables help predict the gold risk premium during recessions. An economic evaluation accounting for transaction costs shows that investors using machine learning methods to estimate expected returns on gold should anticipate limited portfolio gains.</p></div>","PeriodicalId":45111,"journal":{"name":"Journal of Commodity Markets","volume":"31 ","pages":"Article 100293"},"PeriodicalIF":3.7000,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Commodity Markets","FirstCategoryId":"96","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2405851322000502","RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BUSINESS, FINANCE","Score":null,"Total":0}
引用次数: 0
Abstract
This paper assesses the accuracy of several machine learning models’ predictions of the gold risk premium when using an extensive set of 186 predictors. We perform an out-of-sample evaluation and consider both statistical and portfolio metrics. Our results show that machine learning methods and forecast combinations have a limited ability to outperform the historical mean when predicting the gold risk premium. Slightly better results are obtained when predictors are used individually. More specifically, we find that several technical indicators (moving average and momentum series) have forecasting power during periods of expansion, while several business cycle variables and geopolitical risk variables help predict the gold risk premium during recessions. An economic evaluation accounting for transaction costs shows that investors using machine learning methods to estimate expected returns on gold should anticipate limited portfolio gains.
期刊介绍：
The purpose of the journal is also to stimulate international dialog among academics, industry participants, traders, investors, and policymakers with mutual interests in commodity markets. The mandate for the journal is to present ongoing work within commodity economics and finance. Topics can be related to financialization of commodity markets; pricing, hedging, and risk analysis of commodity derivatives; risk premia in commodity markets; real option analysis for commodity project investment and production; portfolio allocation including commodities; forecasting in commodity markets; corporate finance for commodity-exposed corporations; econometric/statistical analysis of commodity markets; organization of commodity markets; regulation of commodity markets; local and global commodity trading; and commodity supply chains. Commodity markets in this context are energy markets (including renewables), metal markets, mineral markets, agricultural markets, livestock and fish markets, markets for weather derivatives, emission markets, shipping markets, water, and related markets. This interdisciplinary and trans-disciplinary journal will cover all commodity markets and is thus relevant for a broad audience. Commodity markets are not only of academic interest but also highly relevant for many practitioners, including asset managers, industrial managers, investment bankers, risk managers, and also policymakers in governments, central banks, and supranational institutions.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
