Early warning systems for financial crises in the conditions of modern Russian economy: specificity and application. Master thesis
Description
The compiled database covers 60 years of recent financial history of 47 economies and is first to include the 2018 crisis in Turkey and the 2020 sovereign debt crisis in Argentina. It covers such variables as debt/gdp levels for different sectors of the economy (state, households, nonfinancial and financial sector), credit/gdp gap, residential property price statistics, The paper is focused on early warning indicators of financial crises applicable to Russia. Using the stepwise regression approach the author identifies early warning indicators for banking and currency crises in advanced and emerging market economies. The proposed prediction model for banking crisis in Russia and emerging market economies includes credit gap and real residential property price index growth. The author explores the possibility of inclusion of residential property price index into the informational base for the countercyclical capital buffer estimation by the Bank of Russia. An analysis of currency crises indicates that private debt-to-service ratio contains useful information for prediction of currency crisis in Russia and emerging market economies. Compiled data is based on statistics published by Bank for International Settlements, Institute for International Finance and Joint External Debt Hub.
Files
Steps to reproduce
Apply logistic regression or test other classification methods.