Elsevier

Resources Policy

Volume 57, August 2018, Pages 196-212
Resources Policy

Gold futures returns and realized moments: A forecasting experiment using a quantile-boosting approach

https://doi.org/10.1016/j.resourpol.2018.03.004Get rights and content

Highlights

  • Quantile boosting sheds light on predictors of gold futures returns.

  • The approach accounts for model uncertainty and model instability.

  • Realized moments have predictive power at intermediate forecast horizons.

  • Realized moments have predictive power in times of distressed markets.

Abstract

This paper proposes an iterative model-building approach known as quantile boosting to trace out the predictive value of realized volatility and skewness for gold futures returns. Controlling for several widely studied market- and sentiment-based variables, we examine the predictive value of realized moments across alternative forecast horizons and across the quantiles of the conditional distribution of gold futures returns. We find that the realized moments often significantly improve the predictive value of the estimated forecasting models at intermediate forecast horizons and across quantiles representing distressed market conditions. We argue that realized moments carry information that reflects investors’ tradeoff between diversification and skewed payoffs, particularly during periods of market stress, which may be especially relevant for gold as the traditional accepted safe haven.

Introduction

The financial market crises and prolonged uncertainty surrounding global economic fundamentals have drawn the attention of researchers towards the dynamics of gold returns as the traditionally accepted safe haven. While the literature has promoted gold as an investment asset due to its low level of correlation with equity indices (Hillier et al., 2006) and its counter-cyclical reaction to unexpected macroeconomic news (Roache and Rossi, 2009), other studies have focused on the determinants of gold returns in the context of its value as a hedge and/or diversifier for investors (Baur and Lucey, 2010, Ciner et al., 2013). Classic determinants of gold returns that have been discussed in earlier literature include stock market returns, exchange-rate movements, oil-price fluctuations, and interest rates (see, for example, Hammoudeh and Yuan, 2008, Pukthuanthong and Roll, 2011, Reboredo, 2013a, 2013b; Pierdzioch et al., 2014, Beckmann et al., 2015), while several recent studies have also pointed out the causal interactions across the gold, oil, and currency markets (e.g. Mo et al., 2018). We contribute to this literature by examining the predictive value of realized moments for gold futures returns across alternative forecast horizons, after controlling for well documented market-based variables and widely studied measures of investor sentiment and uncertainty (Da et al., 2015, Baker et al., 2016). More specifically, we trace out the incremental predictive value of realized volatility and skewness for gold futures returns using a recursively estimated quantile-boosting approach recently used in the literature on gold-price dynamics by Pierdzioch et al. (2016a). The approach accounts for model uncertainty as well as model instability and allows the predictive value of realized moments to be analyzed across the quantiles of the conditional distribution of gold futures returns that represent different market conditions. By doing so, the analysis provides new insights to the predictability of gold futures return that can be useful in hedging and safe-haven analyses.

Driven by the renewed interest in safe haven assets, particularly following the 2007/08 financial crisis, a number of papers have been published in recent years examining the diversification benefits of gold investments. Traditionally, gold has been studied as a hedge against inflation and currency depreciation (e.g., Christie-David et al., 2000, Capie et al., 2005, Worthington and Pahlavani, 2007, Blose, 2010). Following the market turmoil experienced during the credit crunch of 2008, a number of recent studies have also looked into the diversification and safe-haven benefits of gold for stock and bond portfolios (e.g., Baur and Lucey, 2010, Baur and McDermott, 2010, Hood and Malik, 2013, Bredin et al., 2015). While recent studies present mixed evidence in regards to the dominance of gold as a safe haven compared to other assets (e.g., Hood and Malik, 2013, Agyei-Ampomah et al., 2014), the literature generally suggests that gold can work as a hedge and/or safe haven for stock and bond investors both in the U.S. and in other developed markets. At the same time, in another strand of the literature, Lau et al. (2017) examine the recently introduced ETFs for precious metals and document significant return spillovers and volatility transmission across gold, oil, and precious metals with gold ETFs noted as the most influential market. On the other hand, Bams et al. (2017) show that gold price uncertainty is an asset-specific factor that is neither priced across nor within industries, suggesting that the effect of gold as a risk transmitter does not necessarily lead to a risk premium in the cross-section of returns. To that end, modeling gold returns and identifying the market variables that might have predictive value for gold-price fluctuations is of practical concern for hedgers and portfolio managers in the implementation of dynamic diversification and/or hedging strategies. Identification of primary determinants of gold returns can also help enlarge our understanding of volatility transmissions between gold and other market segments, which can especially be useful in cross-hedging strategies. Finally, the findings for gold return dynamics can be employed in further forecasting exercises given the recent evidence that gold prices can help forecast real exchange rates, particularly in the case of major commodity exporters (e.g., Apergis, 2014).

A significant advantage of the quantile-boosting approach utilized in this study is that it follows an iterative model-building procedure in which the forecasting model is built from alternative competing predictor variables. In addition to some of the well-documented market-based variables as well as investor sentiment and uncertainty indicators, we also examine the predictive value of higher-order moments measured by the realized volatility and realized skewness, which we compute using intraday return data. The idea to extend the set of predictors to higher moments is also supported by Baur et al. (2016) who suggest that systematic factors that relate to stock and commodity prices are only partly important in gold price forecasting. In our case, we are particularly interested in the predictive ability of realized volatility and realized skewness as recent research documents that higher-order moments may contain significant information regarding future returns and volatility in stock markets. While studies including Bollerslev et al. (2013) and Corsi et al. (2013) use realized volatility for forecasting stock-market returns and to develop option-valuation models, a number of studies in the asset-pricing literature underline the predictive ability of realized skewness for stock returns. Earlier studies including Barberis and Huang (2008), Brunnermeier et al., 2007, Mitton and Vorkink, 2007 and Boyer et al. (2010) suggest a link between the skewness of individual securities and investors’ portfolio decisions, while Bali et al. (2008) utilize skewness in Value at Risk estimations. In cross-sectional tests, Xing et al. (2010) find that portfolios sorted on a measure related to idiosyncratic skewness generate differences in returns while studies including Barberis and Huang, 2008, Conrad et al., 2013 and Amaya et al. (2015) show that realized skewness has predictive value over subsequent returns. More recently, Kräussl et al. (2016) associate skewness with crash risk, enlarging the scope of risk proxies skewness may be associated with. In a recent application to commodities, Fernandez-Perez et al. (2018) show that a tradeable skewness factor explains the cross-section of commodity futures returns beyond exposures to standard risk premia, while Luo et al. (2016) document that the CBOE Gold ETF Volatility Index and jumps are important factors in forecasting volatility for Shanghai gold futures. Hence, our primary motivation stems from the evidence suggesting that it is important to account for higher-order moments when analyzing return dynamics, and the quantile-boosting approach employed in this study provides an appropriate framework that allows the predictive value of realized moments to be examined in the presence of other well documented market- and sentiment-based indicators as suggested by the recent findings in Balcilar et al., 2017a, Balcilar et al., 2017b.

Our findings suggest that realized moments can significantly improve the predictive value of the estimated forecasting models, even after controlling for widely studied market- and sentiment-based variables. Comparing alternative model specifications that include market-based variables such as the nominal interest rate, term spread, exchange rates, oil and stock market returns as well as popular uncertainty and sentiment indicators, we find that a boosted model that includes realized volatility and skewness often outperforms a simple boosted AR(1) model as well as a boosted model that excludes the realized moments from the list of predictor variables. The predictive value of realized moments is particularly evident for intermediate forecast horizons and holds in many cases for lower quantiles, suggesting that realized moments must be taken into account in forecasting exercises that target distressed market periods in particular. Cross-market tests further suggest that gold return forecasts from a boosted model that includes realized moments have predictive value for the realized moments of stock market returns as well, particularly during periods of high market volatility. These findings are especially important for the estimation of tail-risk measures and Value at Risk projections for periods of market stress and underscore the informational value of realized moments for gold as well as stock market return dynamics. Finally, tests of directional accuracy show that supplementing forecasting models with realized moments can improve the directional accuracy of forecasts for gold returns, which is a matter of particular importance for the implementation of selective hedging strategies in which risk managers base the timing and size of their hedging programs on future price movements.

We organize the remainder of the paper as follows. In Section 2, we explain the quantile-boosting approach and how we evaluate the accuracy of forecasts. In Section 3, we present the data, and we explain how we compute the realized moments. In Section 4, we summarize our empirical results. In Section 5, we conclude with a discussion of practical implications of our empirical results.

Section snippets

The quantile-boosting approach

Researchers have applied quantile-regression techniques in recent research to study several important aspects of gold-price fluctuations (Baur, 2013, Zagaglia and Marzo, 2013) and in a forecasting context (Meligkotsidou et al., 2014). Pierdzioch et al. (2016a) use the quantile-boosting approach to study the statistical accuracy of out-of-sample forecasts of gold returns without examining the predictive value of realized volatility and realized skewness on gold futures returns. The

Gold futures returns and realized moments

Gold futures are traded in NYMEX over a 24 h trading day (pit and electronic). We focus on gold futures prices, rather than spot prices, due to the low transaction costs associated with futures trading, which makes the analysis more relevant for practical applications in the context of hedging and/or safe-haven analyses. Furthermore, as Shrestha (2014) notes, one can expect price discovery to take place primarily in the futures market as the futures price responds to new information faster than

Structure of the forecasting models

For computing our baseline results, we use 75% of the data (1222 observations; the initialization period ends in July 2009; as a robustness check, we shall present results for an extended out-of-sample period in Section 4.6) to initialize the quantile-boosting approach, and the remaining data for out-of-sample forecasting. In total, we have 408 out-of-sample forecasts for every forecast horizon and quantile.

Fig. 2 summarizes the number of iterations it takes to find the minimum of the empirical

Concluding remarks

We have used a recursively estimated quantile-boosting approach to show that realized moments obtained from intraday gold returns have incremental predictive value for gold futures returns over and above several well-documented market-based variables as well as investor sentiment and uncertainty indicators. A significant advantage of the recursively estimated quantile-boosting approach is that it accounts for model uncertainty and model instability, and that it allows the predictive value of

Acknowledgement

We thank an anonymous reviewer for helpful comments. The usual disclaimer applies. CP thanks the German Science Foundation for financial support (Project Macroeconomic Forecasting in Great Crises; Grant number: FR 2677/4-1). The usual disclaimer applies.

References (84)

  • D.G. Baur et al.

    Is gold a safe haven? International evidence

    J. Bank. Financ.

    (2010)
  • J. Beckmann et al.

    Does gold act as a hedge or a safe haven for stocks? A smooth transition approach

    Econ. Model.

    (2015)
  • R. Bhar et al.

    Commodities and financial variables: analyzing relationships in a changing regime environment

    Int. Rev. Econ. Financ.

    (2011)
  • J. Bialkowski et al.

    The gold price in times of crisis

    Int. Rev. Financ. Anal.

    (2015)
  • L.E. Blose

    Gold price, cost of carry, and expected inflation

    J. Econ. Bus.

    (2010)
  • T. Bollerslev et al.

    Risk and return: long run relations, fractional cointegration and return predictability

    J. Financ. Econ.

    (2013)
  • D. Bredin et al.

    Does gold glitter in the long-run? Gold as a hedge and safe haven across time and investment horizon

    Int. Rev. Financ. Anal.

    (2015)
  • F. Capie et al.

    Gold as a hedge against the dollar

    J. Int. Financ. Mark. Inst. Money

    (2005)
  • W.-H. Cheng et al.

    Skewness and leptokurtosis in GARCH-typed VaR estimation of petroleum and metal asset returns

    J. Empir. Financ.

    (2011)
  • R. Christie-David et al.

    Do macroeconomics news releases affect gold and silver prices

    J. Econ. Bus.

    (2000)
  • C. Ciner et al.

    Hedges and safe havens: an examination of stocks, bonds, gold, oil and exchange rates

    Int. Rev. Financ. Anal.

    (2013)
  • F. Corsi et al.

    Realizing smiles: options pricing with realized volatility

    J. Financ. Econ.

    (2013)
  • A. Fernandez-Perez et al.

    The skewness of commodity futures returns

    J. Bank. Financ.

    (2018)
  • I. Figuerola-Ferretti et al.

    The shine of precious metals around the global financial crisis

    J. Empir. Financ.

    (2016)
  • M. Greiner et al.

    Principles and practical application of the receiver-operating characteristic analysis for diagnostic tests

    Prev. Vet. Med.

    (2000)
  • R. Gupta et al.

    Do terror attacks predict gold returns? Evidence from a quantile-predictive-regression approach

    Q. Rev. Econ. Financ.

    (2017)
  • S. Hammoudeh et al.

    Metal volatility in presence of oil and interest rate shocks

    Energy Econ.

    (2008)
  • M. Hood et al.

    Is gold the best hedge and a safe haven under changing stock market volatility?

    Rev. Financ. Econ.

    (2013)
  • R. Kräussl et al.

    Euro crash risk

    J. Empir. Financ.

    (2016)
  • X. Luo et al.

    The information content of implied volatility and jumps in forecasting volatility: evidence from the Shanghai gold futures market

    Financ. Res. Lett.

    (2016)
  • B. Mo et al.

    Dynamic linkages among the gold market, US dollar and crude oil market

    Physica A: Stat. Mech. Appl.

    (2018)
  • C. Pierdzioch et al.

    On the efficiency of the gold market: results of a real-time forecasting approach

    Int. Rev. Financ. Anal.

    (2014)
  • C. Pierdzioch et al.

    A real-time quantile-regression approach to forecasting gold-price fluctuations under asymmetric loss

    Resour. Policy

    (2015)
  • C. Pierdzioch et al.

    A quantile-boosting approach to forecasting gold returns

    North Am. J. Econ. Financ.

    (2016)
  • C. Pierdzioch et al.

    Are precious metals a hedge against exchange-rate movements? An empirical exploration using Bayesian additive regression trees

    North Am. J. Econ. Financ.

    (2016)
  • C. Pierdzioch et al.

    A boosting approach to forecasting the volatility of gold-price fluctuations under flexible loss

    Resour. Policy

    (2016)
  • K. Pukthuanthong et al.

    Gold and the dollar (and the Euro, Pound, and Yen)

    J. Bank. Financ.

    (2011)
  • D.E. Rapach et al.

    Macro variables and international stock return predictability

    Int. J. Forecast.

    (2005)
  • J.C. Reboredo

    Is gold a hedge or safe haven against oil price movements?

    Resour. Policy

    (2013)
  • J.C. Reboredo

    Is gold a safe haven or a hedge for the US dollar? Implications for risk management

    J. Bank. Financ.

    (2013)
  • K. Shrestha

    Price discovery in energy markets

    Energy Econ.

    (2014)
  • Y.-J. Zhang et al.

    The crude oil market and the gold market: evidence for cointegration, causality and price discovery

    Resour. Policy

    (2010)
  • Cited by (12)

    • Rare disaster risks and gold over 700 years: Evidence from nonparametric quantile regressions

      2022, Resources Policy
      Citation Excerpt :

      Our result tend to suggest that, while negative output effects can indeed increase real gold returns, but for that to happen the market must be soaring already, i.e., financial markets are possibly already in turmoil. This is likely to happen if gold prices and or returns are also affected by relatively faster moving behavioral variables, such as economic sentiment, evidence of which can be found in Balcilar et al. (2017) and Bonato et al. (2018), before a rare disaster event fully manifests. In other words, as output starts declining, and that period can be categorized as a rare disaster, negative sentiments are likely to have already set-in to adversely impact the conventional asset markets (see, Da et al. (2015) for a detailed discussion on the link between sentiment and asset prices), and drive gold prices and or returns higher.

    • Cryptocurrency policy uncertainty and gold return forecasting: A dynamic Occam's window approach

      2022, Finance Research Letters
      Citation Excerpt :

      Thus, forecasting the price fluctuations in the gold market and identifying its potential drivers have attracted significant attentions from policymakers, investors, and academics (Alameer et al., 2019; Bonato et al., 2018; Bouri et al., 2021; Chandar et al., 2016; Chiang, 2022; Fang et al., 2018; Gkillas et al., 2020; Li et al., 2021a, 2021b; Livieris et al., 2020; Luo et al., 2022; Madziwa et al., 2022; Wei et al., 2020). The current studies in forecasting gold market can be divided into two major fields, one of which focuses on the forecasting approaches to optimize the predictive accuracy of gold price/return or its volatility (Alameer et al., 2019; Bonato et al., 2018; Chandar et al., 2016; Chen and Xu, 2019; Li et al., 2021b; Madziwa et al., 2022; Risse, 2019), and the other one concentrates on identifying the possible drivers of gold market (Aye et al., 2015; Fang et al., 2018; Gkillas et al., 2020; Hassani et al., 2015; Luo et al., 2022; Mihaylov et al., 2015; Wei et al., 2020). Specifically, classic econometric models and many newly developed machine learning algorithms are widely discussed in the first stream of studies.

    • The (Asymmetric) effect of El Niño and La Niña on gold and silver prices in a GVAR model

      2022, Resources Policy
      Citation Excerpt :

      The GVAR approach is the ideal model for analysing the “true impact” of a global shock such as that of the SOI on real gold and silver returns, as it simultaneously accommodates country-specific macroeconomic conditions and their interlinkages, thereby circumventing omitted variable bias and any potential endogeneity bias resulting from the use of a single-equation model which runs the risk of overestimating the impact of the shock. This is more so, because the literature on predicting gold and silver price movements highlights the role of a wide-array of economic and financial variables of multiple countries (see for example, Pierdzioch et al. (2014a, b, 2015a, b, 2016), Aye et al. (2015), Gupta et al. (2017), Bonato et al. (2018), Dichtl (2020), Pierdzioch and Risse (2020)), which in turn can be modelled via the GVAR. To the best of our knowledge, our current paper is the first to examine the inflation-hedging property of gold and silver in the wake of ENSO cycles based on a GVAR model.

    • Hedging and safe-haven characteristics of Gold against currencies: An investigation based on multivariate dynamic copula theory

      2020, Resources Policy
      Citation Excerpt :

      Alameer et al. (2019) construct a new model to forecast gold price fluctuations based on an improved multilayer perceptron neural network. Bonato et al. (2018) resort to quantile boosting to trace out the predictive value of realized volatility and skewness for gold futures returns. Hernandez et al. (2018) model the return distribution of oil concerning some of the most traded agricultural and precious metal commodities using the cross-quantilogram approach.

    View all citing articles on Scopus
    View full text