This paper, exploiting the properties of mixed causal and noncausal models, proposes strategies to detect time reversibility in stationary stochastic processes. We show that they can also be used for nonstationary processes when the trend component is computed using the Hodrick-Prescott filter characterized by a time-reversible closed form solution. We then establish a linkage between the concept of environmental tipping point and the statistical property of time ir- reversibility and investigate nine climate indicators. While we detect time ir- reversibility in GHG emissions, global temperatures and fundamental natural oscillations do not show this feature. Under a constructive view, our findings give hope that correction policies might still help avoid the worst consequences of climate change.
- c22 - "Single Equation Models; Single Variables: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models"
- mixed causal-noncausal models
- time reversibility
- Global Warming
- generalized Student's t-distribution
- Hodrick-Prescott filter