TY - JOUR ID - 77109 TI - A Note on Early Warning Systems for Monitoring the Inflation of Iran JO - Journal of Algorithms and Computation JA - JAC LA - en SN - 2476-2776 AU - Daadmehr, Elham AU - Habibi, Reza AD - Department of Statistics, Central Bank of Iran AD - Iran Banking Institute, Central Bank of Iran Y1 - 2020 PY - 2020 VL - 52 IS - 1 SP - 183 EP - 195 KW - economic crisis KW - EWS KW - MS model KW - Logistic regression DO - 10.22059/jac.2020.77109 N2 - To check the financial stability, it is important to alarm the possibility of future potential financial crisis. In the literature, the early warning system (EWS) is designed to warn the occurrence of a financial crisis before it happens. This tool gives strengthens to managers to make efficient policy in real economic activities. Hyperinflation, as a financial crisis, is an uncommon bad phenomenon in every economy.  It quickly erodes the real value of the local currency, as the prices of all goods increase. This causes people to minimize their holdings in that currency as they usually switch to more stable foreign currencies, often the US Dollar. Hence, designing a EWS for detecting hyperinflation is valuable task. In the current paper, Iran monthly inflation is modeled by a first  orders autoregressive and moving average model (ARMA) with two-state Markov switching (MS) states, i.e., \( MS \left( 2 \right) -ARMA \left( 1,1 \right) \) . Based on this model, a logistic-EWS is proposed. From the empirical results, it is seen that, in Iran, the low inflation state is more probable than state of high inflation. Beside this, the time of remaining in the low inflation position is almost 9 times more than of high inflation position. To check validity of the results and control prediction errors,it is seen that at least 89 percentages of future states of inflation are correctly predicted with a low noise-to-signal ratio discrepancy measure. UR - https://jac.ut.ac.ir/article_77109.html L1 - https://jac.ut.ac.ir/article_77109_cdcb6d1837eaae2aa6d37e69de5a14d0.pdf ER -