Skip to Main content Skip to Navigation

Revisiting the competitive storage model as a tool for the empirical analysis of commodity price volatility.

Abstract : This thesis proposes an empirical and theoretical analysis of commodity price volatility using the competitive storage model with rational expectations. In essence, the underlying storage theory states that commodity prices are likely to spike when inventory levels are low and cannot buffer the market from exogenous shocks. The prime objective pursued in this dissertation is to use statistical tools to confront the storage model with the data in an attempt to gauge the empirical merit of the storage theory, identify its potential flaws and provide possible remedies for improving its explanatory power. In this respect, the variety of econometric strategies employed so far to test the model itself or its theoretical predictions are reviewed in the opening survey (Ch. 2). The subsequent chapters explore three different routes with the aim of increasing the empirical relevance of the storage framework. Chapter 3 rests on the idea that there might exist long-term movements in the raw commodity price series which have nothing to do with the storage theory. This tends to be confirmed by the results obtained by implementing a hybrid estimation method for recovering jointly the model’s deep parameters with those characterizing the trend. In chapter 4 the testing of the storage theory is pushed even further thanks to the development of an empirical strategy to take the storage model to the data on both prices and quantities, for the first time in the literature. Another novelty is that Bayesian methods are used for inference in contrast to the frequentist approaches employed thus far. Hopefully both these innovations should help paving the way for future research in allowing for the estimation of more complex model set-ups. The last chapter is more theoretical as it deals with the storage model extension on the supply side to account for the dynamics of capital accumulation. The key finding is the crowding-out effect of storage on investment.
Document type :
Complete list of metadata

Cited literature [167 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Friday, May 15, 2020 - 3:22:29 PM
Last modification on : Friday, August 5, 2022 - 2:38:10 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02591704, version 1


Nicolas Legrand. Revisiting the competitive storage model as a tool for the empirical analysis of commodity price volatility.. Economics and Finance. Université Paris Saclay (COmUE), 2016. English. ⟨NNT : 2016SACLA012⟩. ⟨tel-02591704⟩



Record views


Files downloads