ارشیا مهرپور اسکو آبادی
عنوان پایاننامه
مدل سازی چرخه مالی ایران: رهیافت انتقال رژیم مارکوف
- رشته تحصیلی
- علوم اقتصادي-گرايش اقتصاد نظري
- مقطع تحصیلی
- کارشناسی ارشد
- ساعت دفاع
- چکیده
- Abstract Objective: This study aims to model the financial cycle of Iran and evaluate its cyclical asymmetries, including duration, deepness, steepness, and sharpness. The main research questions are which dimensions of financial cycle asymmetry can be reliably identified and tested, and which financial variable contributes most significantly to the composite Financial Conditions Index (FCI). Methodology: The study adopts an integrated empirical framework. First, the core financial variables—broad money (M۲), housing price index, free-market exchange rate, monthly inflation, gold price, interest rate, and the Tehran Stock Exchange Total Index (TEDPIX)—are collected and harmonized at a monthly frequency for the period ۲۰۰۴–۲۰۲۴. A composite Financial Conditions Index is then constructed using a Dynamic Factor Model (DFM) formulated in a state-space framework and estimated via the Kalman filter. The cyclical component of the index is extracted using the Christiano–Fitzgerald band-pass filter. Subsequently, the financial cycle dynamics are modeled through a two-regime Markov-Switching Autoregressive (MS-AR) model to estimate regime-specific means, variances, and transition probabilities, and to date episodes of financial easing and tightening. Duration, deepness, steepness, and sharpness asymmetries are evaluated using Wald tests, while robustness is assessed through the Hodrick–Prescott (HP) filter and unit root tests. Conclusion: The proposed framework standardizes and enhances the interpretability of Iran's financial cycle by coherently aggregating information from key financial variables and separating common cyclical movements from idiosyncratic shocks. The MS-AR model enables a structural comparison of financial cycle behavior across regimes and provides a systematic framework for testing cyclical asymmetries. Furthermore, the estimated factor loadings identify the dominant financial component in the construction of the composite index, thereby clarifying the principal drivers of the financial cycle. Robustness checks strengthen the reliability of the findings and provide a solid foundation for financial cycle monitoring and macroprudential policy applications, including the development of early warning systems. Keywords: Financial Cycle; Markov-Switching Autoregressive (MS-AR) Model; Cyclical Asymmetry; Dynamic Factor Model; Christiano–Fitzgerald Filter.
- Abstract
- Abstract Objective: This study aims to model the financial cycle of Iran and evaluate its cyclical asymmetries, including duration, deepness, steepness, and sharpness. The main research questions are which dimensions of financial cycle asymmetry can be reliably identified and tested, and which financial variable contributes most significantly to the composite Financial Conditions Index (FCI). Methodology: The study adopts an integrated empirical framework. First, the core financial variables—broad money (M۲), housing price index, free-market exchange rate, monthly inflation, gold price, interest rate, and the Tehran Stock Exchange Total Index (TEDPIX)—are collected and harmonized at a monthly frequency for the period ۲۰۰۴–۲۰۲۴. A composite Financial Conditions Index is then constructed using a Dynamic Factor Model (DFM) formulated in a state-space framework and estimated via the Kalman filter. The cyclical component of the index is extracted using the Christiano–Fitzgerald band-pass filter. Subsequently, the financial cycle dynamics are modeled through a two-regime Markov-Switching Autoregressive (MS-AR) model to estimate regime-specific means, variances, and transition probabilities, and to date episodes of financial easing and tightening. Duration, deepness, steepness, and sharpness asymmetries are evaluated using Wald tests, while robustness is assessed through the Hodrick–Prescott (HP) filter and unit root tests. Conclusion: The proposed framework standardizes and enhances the interpretability of Iran's financial cycle by coherently aggregating information from key financial variables and separating common cyclical movements from idiosyncratic shocks. The MS-AR model enables a structural comparison of financial cycle behavior across regimes and provides a systematic framework for testing cyclical asymmetries. Furthermore, the estimated factor loadings identify the dominant financial component in the construction of the composite index, thereby clarifying the principal drivers of the financial cycle. Robustness checks strengthen the reliability of the findings and provide a solid foundation for financial cycle monitoring and macroprudential policy applications, including the development of early warning systems. Keywords: Financial Cycle; Markov-Switching Autoregressive (MS-AR) Model; Cyclical Asymmetry; Dynamic Factor Model; Christiano–Fitzgerald Filter.