Policy responses to climate change rely on models of both the underlying socio-economic and physical processes, facing unstable observations measured with error, to produce projections of future scenarios. But data to evaluate impacts and establish historical climate records are non-stationary: distributions shift over time due to shocks, measurement changes and stochastic trends – all of which invalidate standard statistical inference.

The project develops econometric methods to advance climate research, by disentangling the complex relationships between human actions and climate responses masked by stochastic trends and breaks.

In order to achieve this, Dr Pretis will firstly establish econometric methods to model the non-stationary nature of climate data consistent with known physical laws, enabling joint estimation and testing. He will then develop techniques for automatic detection of structural shifts to evaluate the impacts of policy and improve historical records. Finally, he will apply the resulting methods to improve socio-economic projections in climate predictions, as used by the Intergovernmental Panel on Climate Change.


Funded by: British Academy

Duration: 3 years

Related Groups