skip to content

Keynes Fund

 

Project Summary


The goal of the project is to develop econometric techniques to study climate change and forecast its consequences. We interpret the subject of climate widely to include environmental issues in general and entitle our project Persistence and Forecasting in Climate and Environmental Econometrics.

We will structure our project into the development of statistical techniques to analyse climatological time series and the building of a forecasting model for policy-relevant environmental outcomes. A new method that offers a representation of climate policy-relevant parameters as well as an estimation and inference procedure will, therefore, be at the heart of our collaboration. Furthermore, we shall generate forecasts of quantities such as temperature changes which economists may use to construct counterfactuals and projections. We will develop software in the form of an R package and a Stata command along with any theoretical results, parameter estimates based on past data, and forecasts based on our results to cause widespread impact.

Econometricians’ techniques from the last fifty years are very different from climate scientists’ approaches. Hence, there is potential for making important contributions to knowledge that might prove useful in our understanding and prediction of climatic and environmental data both in their own right and to inform empirical studies of economic consequences. Econometricians at Oxford have a dedicated research group on this topic and in the last four years there has been an annual international conference on Econometric Models of Climate Change.

Both of us aim to collaborate closely with the Oxford group, in particular with James Duffy and other econometricians at Oxford, who have already confirmed their interest to become involved with this project.

A good example of the application of econometric knowledge to environmental science is the time series model developed by Harvey et al. (2019) for circular data, such as wind direction. Its main inspiration originates from recently developed techniques for the prediction of volatility in financial time series, as developed in Harvey (2013), Creal et al. (2013), Harvey and Lange (2017), and Harvey and Palumbo (2019), which use derivative information of the statistical objective function to explain observed dynamics. Application of these econometric results solved an open statistical problem in environmental science. AH presented a paper on modelling directional time series at the last Climate Econometrics conference, held in Milan in August 2019.

Similarly, the leading application of our statistical results concerns the relationship between radiative forcing, Earth’s ability to trap heat, and global average temperature deviation as shown in Figure 1. Kaufmann et al. (2013) provide statistical and physical evidence that neither series has a time-invariant distribution, which renders econometric approaches amenable to this problem.

Anthropogenic climate change and adverse ramifications perhaps present the greatest market failure of all time. Its statistical study and an estimate of its economic consequences therefore lie within the aims and objectives of the Keynes fund. Indeed, both sources and consequences of climate change originate from failure of market efficiency: the actions or inaction of the generations currently alive generate agency costs that future generations have to bear if climate change causes livelihoods to disintegrate.

 

 

Prof. Andrew Harvey

 

Professor Andrew Harvey is Emeritus Professor of Econometrics at the Faculty of Economics, University of Cambridge and Fellow of the Econometric Society and a Fellow of the British Academy (FBA). His research expertise is in Time Series, Financial Econometrics, State Space Models, Signal Extraction, Volatility.

 

Share this Project