Il giorno 8 febbraio alle ore 12 presso l'Aula 3 del Dipartimento di Matematica e Informatica, il Professor Espen Sirnes (Sirnes, Espen | UiT), docente della School of Business and Economics di Tromsø in Norvegia, terrà un seminario dal titlo "Paneltime, a python package for analyzing panel data with timeseries problems".
Paneltime is a python package developed for analyzing large panel datasets with timeseries problems. The package was developed while working on financial data, but it is quite common to encounter such problems in all sorts of data. The package handles panel issues (fixed/random effects) and timeseries issues (GARCH/ARIMA) simultaneously. The hessian is calculated analytically, which increases efficiency, accuracy and the ability to handle multicollinearity. The package uses multicore calculation extensively, with each node starting on different initial arguments.
The project is available on GitHub: https://github.com/espensirnes/paneltime