El Niño, La Niña, and forecastability of the realized variance of agricultural commodity prices: Evidence from a machine learning approach

Abstract

We examine the predictive value of El Niño and La Niña weather episodes for the subsequent realized variance of 16 agricultural commodity prices. To this end, we use high‐frequency data covering the period from 2009 to 2020 to estimate the realized variance along realized skewness, realized kurtosis, realized jumps, and realized upside and downside tail risks as control variables. Accounting for the impact of the control variables as well as spillover effects from the realized variances of the other agricultural commodities in our sample, we estimate an extended heterogeneous autoregressive (HAR) model by means of random forests to capture in a purely data‐driven way potentially nonlinear links between El Niño and La Niña and the subsequent realized variance. We document such nonlinear links, and that El Niño and La Niña increase forecast accuracy, especially at longer forecast horizons, for several of the agricultural commodities that we study in this research.

Climate Risks and Stock Market Volatility over a Century in an Emerging Market Economy: The Case of South Africa

“…In fact, in earlier studies on climate risks and stock markets, researchers primarily have concentrated on developed countries and on in-sample movements of the first moment [13,15,27,29,30], with the only exception being [14], who have analyzed stock market volatility of the state-level data in the United States (US). When it comes to volatility, the literature thus far has concentrated on predicting second moments of commodity returns due to climate risks (for example, [58][59][60][61][62][63]). Another somewhat related paper is that of [64], who have forecasted indicators of financial stress, comprised of both first and second moments of the underlying assets, of developed countries.…”

Forecasting the realized volatility of agricultural commodity prices: Does sentiment matter?

“…The validation sample, , provides new predictions of , which we combine with the predictions from the previous iteration to compute a vector of average predicted realized volatilities. As we proceed from one iteration to the next and average predictions from an increasing number of random HAR‐RV‐Sentiment‐Moments models, we eventually obtain a “stable” vector of average predicted realized volatilities. We terminate the iterative process once the maximum absolute percentage change in the vector of predicted realized volatilities becomes sufficiently small (like Bonato et al, 2022, we use the threshold 0.01 to operationalize “sufficiently small”).…”

“…While our primary focus is on investigating the role of investor sentiment in forecasting the of multiple agricultural commodities price returns, it is also essential to compare the performance of sentiment with that of realized moments. The literature on forecasting of agricultural commodities price returns has emphasized the importance of realized moments, such as leverage, realized skewness, realized kurtosis, realized upside volatility, realized downside volatility, realized jumps, realized upside tail risk, and realized downside tail risk (see, e.g., Bonato et al, 2022; Chatziantoniou et al, 2021; Degiannakis et al, 2022; Luo et al, 2019; Marfatia et al, 2022; Shiba et al, 2022; Tian et al, 2017a, 2017b; Yang et al, 2017). Given that we consider several realized moments as candidates for forecasting , we construct HAR‐RV‐Sentiment‐Moments forecasting models by two alternative approaches: the forward and backward stepwise predictor selection algorithm (see the textbook by Hastie et al, 2009) and a Model‐Based Averaging (MOBA) algorithm (Bonato et al, 2023).…”

Climate Risks and Forecasting Stock Market Returns in Advanced Economies over a Century

“…Finally, another interesting avenue for future research is to switch from the kind of out-of-sample analysis that we have undertaken in this research to the type of out-of-bag analysis often used in the machinelearning literature. For a recent application of an out-of-bag analysis in empirical finance, see [65].…”