-
Notifications
You must be signed in to change notification settings - Fork 2
A methodology based on Markov chains and dynamic entropy measures is proposed for measuring and forecasting the evolution of the inequality of financial risks
License
lstorchi/markovctheil
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
Minimal version of Forecasted dynamic Theil's entropy software: To run the program you need Python 3, PyQt5, matplotlib, numpy and scipy. The code has been tested and developed under Linux, but once installed the needed packages should work also under Mac OS X and Windows OS. To run the GUI: $ python randentropy_qt.py to run the GUI you may need to export QT_X11_NO_MITSHM=1 to run the CLI: $ python randentropy.py Finally you'll find also a changepoint.py CLI Quick start: to perform some test using the changepoint CLI you can use the input files in ./files specifically: $ python3 changepoint.py -m ./files/sep_monthly.mat -c 1 you should get Change Point: 158 ( -320.0611462871186 ) as a result. Similarly you can run the same using the GUI. To use two change-points: $ python3 changepoint.py -m ./files/sep_monthly.mat -c 2 \ --cp2-start 50 --cp2-stop 157 you should get the following results: Change Point: 73 , 119 ( -311.1237648393643 ) You can run also the Lambda test: $ python3 changepoint.py -m ./files/sep_monthly.mat -c 2 \ --perform-test "73;119;100" To test the randentropy CLI, you can use the same file as before: $ python3 randentropy.py -m ./files/sep_monthly.mat \ -b ./files/sep_monthly.mat -s 0.25 -t 36 -n 1000 -v the sep_monthly.mat contains both the community as well as the attributes matrices. The same results (i.e. entropy_1000.txt in this case) can be obtained using the GUI (i.e. randentropy_qt.py). Cite this as: D’Amico G, Scocchera S, Storchi L (2018b). “Financial risk distribution in European Union.” Physica A: Statistical Mechanics and its Applications, 505, 252–267. D’Amico G, Petroni F, Regnault P, Scocchera S, Storchi L (2019). “A Copula-based Markov Reward Approach to the Credit Spread in the European Union.” Applied Mathematical Finance, 26(4), 359–386. D’Amico G, Regnault P, Scocchera S, Storchi L (2018a). “A Continuous-Time InequalityMeasure Applied to Financial Risk: The Case of the European Union.” International Journal of Financial Studies, 6(3), 62.
About
A methodology based on Markov chains and dynamic entropy measures is proposed for measuring and forecasting the evolution of the inequality of financial risks
Resources
License
Stars
Watchers
Forks
Packages 0
No packages published