We propose a novel methodology for forecasting chaotic systems which is based on exploiting the information conveyed by the local Lyapunov exponents of a system. This information is used to correct for the inevitable bias of most non-parametric predictors. Using simulated data, we show that gains in prediction accuracy can be substantial.

Due to meteorological factors, the distribution of the environmental damage due to climate change bears no relationship to that of global emissions. We argue in favor of o¤setting this discrepancy, and propose a "global insurance scheme"to be …ncanced according to countries'responsibility for climate change. Because GHG decay very slowly, we argue that the actual burden of global warming should be shared on the basis of cumulated emissions, raher than sharing the expected costs of actual emissions as in a Pigovian taxation scheme. We characterize new versions of two well-known cost-sharing schemes by adapting the responsibility theory of Bossert and Fleurbaey (1996) to a context with externalities.

Climate ethics have been concerned with polluter pays, beneficiary pays and ability to pay principles, all of which consider climate change as a single negative externality. This paper considers it as a constellation of externalities, positive and negative, with different associated demands of justice. This is important because explicitly considering positive externalities has not to our knowledge been done in the climate ethics literature. Specifically, it is argued that those who enjoy passive gains from climate change owe gains not to the net losers, but to the emitters, just as the emitters owe compensation to the net losers for the negative externality. This is defended by appeal to theoretical virtues and to the social benefits of generating positive externalities, even when those positive externalities are coupled with far greater negative externalities. We call this the Polluter Pays, Then Receives (‘PPTR', or ‘Peter') Principle.