Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock Returns
This paper used a factor-augmented vector autoregressive model with time-varying coefficients to construct a financial conditions index. Time variation in the model’s parameters allowed the weights to be attached to each variable in the index to evolve and evaluate dynamics across time. The ability...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | fas |
| Published: |
University of Isfahan
2022-03-01
|
| Series: | Journal of Asset Management and Financing |
| Subjects: | |
| Online Access: | https://amf.ui.ac.ir/article_26718_d48a8737c9f2ea8a87ad694abd473dc2.pdf |
| _version_ | 1797806949405818880 |
|---|---|
| author | Seyed Aziz Arman Ebrahim Anvari Samere RakiKianpour |
| author_facet | Seyed Aziz Arman Ebrahim Anvari Samere RakiKianpour |
| author_sort | Seyed Aziz Arman |
| collection | DOAJ |
| description | This paper used a factor-augmented vector autoregressive model with time-varying coefficients to construct a financial conditions index. Time variation in the model’s parameters allowed the weights to be attached to each variable in the index to evolve and evaluate dynamics across time. The ability of the constructed index to predict various variables was also evaluated. The Financial Condition Index (FCI) was estimated by using the TVP-FAVAR method based on the quarterly data of the period of 1989-2019. The variables used included interest rate, exchange rate growth, inflation rate, consumption growth, banking facility growth, total stock market index growth, money supply growth, oil revenue growth, and gross domestic product growth rate. The findings indicated significant volatilities in the model’s parameters. The shock from improving the FCI led to a positive response to the stock market index. According to the findings, the constructed FCI had high predictability.IntroductionThis paper reviewed the Financial Conditions Index (FCI) in the context of Iran. An FCI combines at least 4 financial prices: a short interest rate, a bond rate, an exchange rate, and a stock price index. The mentioned index may have the ability to summarize financial conditions. Therefore, it can be a valuable tool for policymakers, households, and firms. Monetary policymakers can also employ FCI to investigate the extensive effects of monetary policy on financial markets. The construction and use of FCI involve 3 issues, including selection of variables to enter into FCI, weights that are used to average these variables, relationship between FCI and macroeconomy, and assessment of the predictive power of this index for economic variables. This paper used a factor-augmented vector autoregressive model with Time-Varying Parameter Factor-Augmented Vector Auto-Regressive (TVP-FAVAR) coefficients to construct the index. Time variation in the model’s parameters allowed the weights to be attached to each variable in the index to evolve and evaluate dynamics across time. Then, the ability of this index to predict various variables, including stock returns, was evaluated. Method and DataThe p-lag TVP-FAVAR model in this paper took the following form: = + + where is the regression coefficient; is factor loading; is the latent factor that can interpret as FCI; is a vector of intercepts; are VAR coefficients; and and are zero-mean Gaussian disturbances with the time-varying covariances of Vt and Qt, respectively. The model was estimated by using the Markov Chain Monte Carlo (MCMC) methods. Short-term-investment deposit rate (one-year), non-official exchange rate growth, inflation rate, consumption growth, banking facility growth, total stock market index growth, money supply growth, oil revenue growth, and GDP growth were selected as the model variables to construct the FCI. The model estimations were made by using the quarterly data of 1989-2019. The data were extracted from the official website of the Central Bank of Iran and the Economic and Financial Databank of Iran. All the series were seasonally adjusted by using the X-12 procedure. FindingsThe augmented Dickey-Fuller (ADF) and Zivot-Andrews unit root tests were performed. All the series were stationary in level or first differences. According to the Bai-Ng criteria, the number of factors was estimated to be two. According to the Schwarz information criterion, the number of lags was estimated to be one. The results indicated significant volatility of the developed FCI index. Nevertheless, the stochastic volatility or variance of the error terms of the financial condition index decreased. The posterior mean results showed that the oil revenue and money supply shocks could positively affect the FCI. The Impulse Response Function (IRF) indicated that the gross domestic product positively responded to the shock in the financial condition index only for a short time, while its effect was negative in the second period. Moreover, the effect of the shock disappeared and in the long run did not affect the GDP. The growth in the consumption, exchange rate growth, and inflation rate positively responded to the FCI shock. Finally, the stock market index growth positively responded to the FCI shock within 10 periods. Predictions of the responses of the variables to the shock based on the financial condition index (4-period ahead, 8-period ahead, and 12-period ahead) indicated the high predictive power of the model. In addition, the results of the in-sample and out-of-sample prediction errors, Root Mean Square Error (RMSE), and Theil’s Inequality Coefficient (TIC) represented the high predictive power of the model. Conclusion and discussion Knowing that a financial condition index could be a useful tool for policymakers, an FCI was developed specifically for Iran. Our results suggested that investors should analyze the government’s previous and future decisions and policies and evaluate the macroeconomic variables before investing in the stock market. In addition, it is suggested that the stock market variable, which is one of the channels of the monetary transmission mechanism, be treated as an active monetary-policy mechanism in Iran although its inefficiency requires further attention. |
| first_indexed | 2024-03-13T06:15:09Z |
| format | Article |
| id | doaj.art-fb2e6c5de1464b3692840d6ba7c82c57 |
| institution | Directory Open Access Journal |
| issn | 2383-1189 |
| language | fas |
| last_indexed | 2024-03-13T06:15:09Z |
| publishDate | 2022-03-01 |
| publisher | University of Isfahan |
| record_format | Article |
| series | Journal of Asset Management and Financing |
| spelling | doaj.art-fb2e6c5de1464b3692840d6ba7c82c572023-06-11T08:53:54ZfasUniversity of IsfahanJournal of Asset Management and Financing2383-11892022-03-01101477210.22108/amf.2022.129138.167226718Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock ReturnsSeyed Aziz Arman0Ebrahim Anvari1Samere RakiKianpour2Professor, Department of Economics, Faculty of Economics and Social Sciences, Shahid Chamran University, Ahvaz, IranAssociate Professor, Department of Economics, Faculty of Economics and Social Sciences, Shahid Chamran University, Ahvaz, IranPh. D. Candidate, Department of Economics, Faculty of Economics and Social Sciences, Shahid Chamran University, Ahvaz, IranThis paper used a factor-augmented vector autoregressive model with time-varying coefficients to construct a financial conditions index. Time variation in the model’s parameters allowed the weights to be attached to each variable in the index to evolve and evaluate dynamics across time. The ability of the constructed index to predict various variables was also evaluated. The Financial Condition Index (FCI) was estimated by using the TVP-FAVAR method based on the quarterly data of the period of 1989-2019. The variables used included interest rate, exchange rate growth, inflation rate, consumption growth, banking facility growth, total stock market index growth, money supply growth, oil revenue growth, and gross domestic product growth rate. The findings indicated significant volatilities in the model’s parameters. The shock from improving the FCI led to a positive response to the stock market index. According to the findings, the constructed FCI had high predictability.IntroductionThis paper reviewed the Financial Conditions Index (FCI) in the context of Iran. An FCI combines at least 4 financial prices: a short interest rate, a bond rate, an exchange rate, and a stock price index. The mentioned index may have the ability to summarize financial conditions. Therefore, it can be a valuable tool for policymakers, households, and firms. Monetary policymakers can also employ FCI to investigate the extensive effects of monetary policy on financial markets. The construction and use of FCI involve 3 issues, including selection of variables to enter into FCI, weights that are used to average these variables, relationship between FCI and macroeconomy, and assessment of the predictive power of this index for economic variables. This paper used a factor-augmented vector autoregressive model with Time-Varying Parameter Factor-Augmented Vector Auto-Regressive (TVP-FAVAR) coefficients to construct the index. Time variation in the model’s parameters allowed the weights to be attached to each variable in the index to evolve and evaluate dynamics across time. Then, the ability of this index to predict various variables, including stock returns, was evaluated. Method and DataThe p-lag TVP-FAVAR model in this paper took the following form: = + + where is the regression coefficient; is factor loading; is the latent factor that can interpret as FCI; is a vector of intercepts; are VAR coefficients; and and are zero-mean Gaussian disturbances with the time-varying covariances of Vt and Qt, respectively. The model was estimated by using the Markov Chain Monte Carlo (MCMC) methods. Short-term-investment deposit rate (one-year), non-official exchange rate growth, inflation rate, consumption growth, banking facility growth, total stock market index growth, money supply growth, oil revenue growth, and GDP growth were selected as the model variables to construct the FCI. The model estimations were made by using the quarterly data of 1989-2019. The data were extracted from the official website of the Central Bank of Iran and the Economic and Financial Databank of Iran. All the series were seasonally adjusted by using the X-12 procedure. FindingsThe augmented Dickey-Fuller (ADF) and Zivot-Andrews unit root tests were performed. All the series were stationary in level or first differences. According to the Bai-Ng criteria, the number of factors was estimated to be two. According to the Schwarz information criterion, the number of lags was estimated to be one. The results indicated significant volatility of the developed FCI index. Nevertheless, the stochastic volatility or variance of the error terms of the financial condition index decreased. The posterior mean results showed that the oil revenue and money supply shocks could positively affect the FCI. The Impulse Response Function (IRF) indicated that the gross domestic product positively responded to the shock in the financial condition index only for a short time, while its effect was negative in the second period. Moreover, the effect of the shock disappeared and in the long run did not affect the GDP. The growth in the consumption, exchange rate growth, and inflation rate positively responded to the FCI shock. Finally, the stock market index growth positively responded to the FCI shock within 10 periods. Predictions of the responses of the variables to the shock based on the financial condition index (4-period ahead, 8-period ahead, and 12-period ahead) indicated the high predictive power of the model. In addition, the results of the in-sample and out-of-sample prediction errors, Root Mean Square Error (RMSE), and Theil’s Inequality Coefficient (TIC) represented the high predictive power of the model. Conclusion and discussion Knowing that a financial condition index could be a useful tool for policymakers, an FCI was developed specifically for Iran. Our results suggested that investors should analyze the government’s previous and future decisions and policies and evaluate the macroeconomic variables before investing in the stock market. In addition, it is suggested that the stock market variable, which is one of the channels of the monetary transmission mechanism, be treated as an active monetary-policy mechanism in Iran although its inefficiency requires further attention.https://amf.ui.ac.ir/article_26718_d48a8737c9f2ea8a87ad694abd473dc2.pdffinancial condition index (fci)forecasting stock returnstime-varying parameter (tvp) model |
| spellingShingle | Seyed Aziz Arman Ebrahim Anvari Samere RakiKianpour Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock Returns Journal of Asset Management and Financing financial condition index (fci) forecasting stock returns time-varying parameter (tvp) model |
| title | Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock Returns |
| title_full | Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock Returns |
| title_fullStr | Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock Returns |
| title_full_unstemmed | Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock Returns |
| title_short | Modeling the Dynamic Financial Condition Index (FCI) and Assessing Its Effectiveness in Predicting Iran’s Stock Returns |
| title_sort | modeling the dynamic financial condition index fci and assessing its effectiveness in predicting iran s stock returns |
| topic | financial condition index (fci) forecasting stock returns time-varying parameter (tvp) model |
| url | https://amf.ui.ac.ir/article_26718_d48a8737c9f2ea8a87ad694abd473dc2.pdf |
| work_keys_str_mv | AT seyedazizarman modelingthedynamicfinancialconditionindexfciandassessingitseffectivenessinpredictingiransstockreturns AT ebrahimanvari modelingthedynamicfinancialconditionindexfciandassessingitseffectivenessinpredictingiransstockreturns AT samererakikianpour modelingthedynamicfinancialconditionindexfciandassessingitseffectivenessinpredictingiransstockreturns |