Jointly supported by Rokko Forum / Kobe University Center for Social System Innovation
|Date & Time||Tuesday, May 09, 2023, 15:10 - 16:40|
|Place||Hybrid (Face to Face / Online by Zoom)|
|Intended Audience||Faculty, Graduate Students, and People with Equivalent Knowledge|
|Registration||Registration is required. The seminar details will be sent to the registered emails.
Registration Form (Due: May 1)
15:10 - 16:40
- Revealing Priors from Posteriors (co-authored with Masako Ikefuji and Takashi Yamagata)
- Jan MAGNUS (School of Business and Economics, Econometrics and Data Science, Vrije Universiteit Amsterdam)
- A Bayesian typically uses data and a prior to produce a posterior. In practice, however, the prior is often not observed while the data and the posterior are. We shall follow the opposite route, using data and the posterior information to reveal the prior. We then apply this theory to inflation forecasts by the Bank of England and to (equilibrium) climate sensitivity as reported by the Intergovernmental Panel on Climate Change in an attempt to get some insight into the prior beliefs of central bankers (especially under the uncertainty about Brexit and the corona crisis) and IPCC scientists. It appears that the IPCC scientists have agreed on a value for the climate equilibrium between 3 and 4 degrees Celsius, while judging the occurrence of a real disaster much more likely than the previous report predicts.