استخدام نموذج الانحدار غير الخطي اللوغاريتم الاحتمالي ثنائي الحد للتنبؤ بأزمة العملة

Abstract EN

The best method to avoid financial crises in economical systems is to predict financial imbalances and correct them using early warning systems before they turn to be a real crisis.

This study aims to test the effectiveness of a non-linear regressive system, which is based on binary logarithmic probability, on predicting currency crisis.

Data used is the one of countries that faced currency crises (Turkey, Indonesia).

The next step is to analyze data signals during and before the crisis, as well as the use of ADF and Sensitivity tests to test the quality of the model.

The results show that the EWS, used in this study, has the capacity to predict currency crises two years before they happen, and the standards tests have proved its efficiency too.