Predicting Financial Instability in Ukraine by Use of Warning Signal Indicators

Abstract

An essential objective of statistical science is quick and timely elaboration of new approaches and models of financial system development, for solving specific ad hoc problems and for statistical grounding of preventive measures taken to avoid or mitigate the impact of negative factors capable to trigger shocks in financial system. Computations of threshold values of statistical indicators by data for 2006-2009 are made by use of the method for selection of leading signal indicators. It allows for testing their prediction capacity, which is the capacity for accurate signaling on the crisis occurrence before or in time of a financial and economic crisis.

The best prediction capacity is shown by the indicators of consumer price index, interest rate at interbank credit market, and balance on current account, which, when used, allow for increasing, by more than 50 percentage points, the probability of predicting the occurrence of financial instability in case of their signaling compared to unconditional probability of the instability Also, the following indicators are effective: exchange rate of the Ukrainian hryvnya, external public debt, surplus of money, capital outflow, dynamics of gold reserves, and ratio of interest on loans to interest on deposits. The set of determined effective indicators is used for predicting the probability of the occurrence of financial crisis in 2015 by evaluating these indicators for 2014.