جستجوی مقالات مرتبط با کلیدواژه « میانگین گیری مدل بیزین » در نشریات گروه « اقتصاد »

The present research aimed to investigate the relationship between Iran’s stock price index and nine macroeconomic variables during 1996–2019. Three methods were employed to reduce uncertainty, namely three Bayesian averaging methods (BMA, BMS, BAS), weighted average least squares (WALS), and Vselect. The experimental results of the Bayesian methods and WALS showed that the exchange rate and the consumer price index are the most important variables among the nine macroeconomic variables considered in the model. Moreover, the results revealed that the exchange rate has a minor impact on the stock price index, while the stock price index exerts a substantial effect on the exchange rate. The findings of Vselect validated the conclusion that these two variables are the primary drivers of stock price estimation and are present in nearly all predictive modelsIntroductionThe harmonization of financial markets with the macroeconomic sector is crucial for stabilizing the economy and achieving the adopted policies. In recent years, several significant studies have been conducted on financial markets, particularly the stock market, highlighting their pivotal role in allocating capital resources efficiently in advanced economies. Empirical evidence supports the view that financial markets have evolved in tandem with all sectors of the economy. Therefore, it can be argued that financial markets constitute one of the most vital components of any country’s economy. Throughout history, major economic crises have resulted from the collapse of financial markets, which underscores their critical significance. The financial market comprises several components, with the stock market being a crucial part. Economists view it as a barometer of a country’s economic health due to its ability to reflect macroeconomic asset prices more accurately than other markets. The uncertainty surrounding stock prices in stock markets is a significant aspect of the entire economy, capable of generating and disrupting unsustainable growth. For investors, the risk of participating in an investment is a crucial consideration. To comprehend total risk, it is beneficial to examine two aspects: systematic and non-systematic risk.The present study aimed to examine the impact of economic factors on stock market prices in Iran with the high degree of risk involved. There is a consensus among economists that asset prices are responsive to economic news, and that stock prices and economic factors are strongly interconnected. Thus, this research investigated the potential impact of macroeconomic factors on the Iranian stock price index from 1996 to 2019 using Bayesian averaging methods, followed by an analysis of the effect size of each variable through the weighted average least square method (WALS).Materials and MethodsResearchers often draw conclusions based on the assumptions of their selected model, assuming that it can accurately predict real-world situations. However, this approach may overlook true uncertainty, leading to non-conservative conclusions. Statistical models comprise two parts: variables and assumptions, and the model selected based on these assumptions to estimate the variables. Uncertainty exists at both levels. For instance, a researcher estimating the impact of influential factors on an independent variable may choose a model based on their assumptions and report their estimates. But is this the best answer? Another researcher with different assumptions may opt for a different model with lower variance and error. In other words, numerous models may fit the sample data equally well but with different coefficient estimates and standard errors. Bayesian model averaging (BMA) is a robust method that aims to remove uncertainty. It assesses the robustness of results to alternative specifications by computing posterior distributions for coefficients and models. This study employed three models of BMA, BMS, and BAS, using various averaging methods to verify the reliability of the results. Moreover, two non-Bayesian methods, namely WALS and Vselect, were used to select the best variables for predicting the optimal models.ConclusionThis study tried to investigate the relationship between Iran’s stock market index and nine macroeconomic variables during 1996–2019 by using the models that identify and limit uncertainty. The models selected include three Bayesian averaging models as well as WALS and Vselect which were used to verify the results obtained. The results indicated that only two variables, the exchange rate and consumer price index, are statistically significant when assuming a uniform distribution of the prior distribution function, which is the assumption of the BMS method. The remaining variables are not statistically significant. Furthermore, the estimates derived from the BMA and BAS models were quite similar, with the exception of less important variables. However, the similarity decreased in the BAS method. Moreover, WALS and Vselect confirmed the results obtained from all the three methods.

Tax evasion is one of the main activities of the informal sector of the economy and is one of the most important issues and problems of the tax system. Given the fundamental role of tax evasion in reducing public government revenues; Awareness of the reasons and factors determining the tax evasion of Iran's economy is important for policymakers and financial officials to control and reduce its growth. Therefore, in the present study, tax evasion during the period 1360-1396 has been estimated using the  new monetary demand function approach and the VECM vector error correction model. Then, using Bayesian econometric approach and using Bayesian model averaging (BMA) method, the factors affecting tax evasion in Iran's economy have been studied. The results of this study show that the trend of tax evasion index has been increasing during the period under review. Also, the variables of tax burden, degree of urbanization, oil revenues, unemployment rate and tax complexity have had the greatest impact on tax evasion.

The present study represents an attempt to examine the main macro determinants of stock prices in OPEC oil exporting and importing countries in the study period 1996 to 2016 using Bayesian model averaging (BMA). The oil importers in this study are the United States, Britain and Japan, and three countries, Iran, Saudi Arabia and Kuwait, have been selected as oil exporters. The findings of this study are that to predict and evaluate the stock price index for oil-importing countries, the three variables of exchange rate index, consumer price index and economic growth should be given more importance than other variables, while for oil-exporting countries, The three variables of broad money growth, exchange rate and import are the most important variables that should be considered. For oil-importing countries, among the macro variables studied, OPEC oil prices have a completely negative relationship with the stock price index of those countries, but in oil-importing countries, the gold price has a completely inverse relationship with the stock price index.

Predictable and unpredictable delays in the national accounts data dissemination in Iran highlight the nowcasting of the economy’s state with using timely and high-frequency data. The large share of service sectors in GDP make forecasting of this sector more important. This paper seeks to answer the question of whether the status of the service and transportation sector can be predicted by using the vehicle traffic count dataset. In this regard, daily data on 2590 points of the country's roads from 2015 to September 2021 is used. In addition to using a simple aggregation method to construct the index, Artificial Neural Network model (ANNs) and Bayesian Model Averaging (BMA) are also used. The results show that the estimation indices extracted from these data have less forecast error than the benchmark models (ARMA) and can represent changes in both services and transportation sectors. The comparison of different methods of index construction shows the index extracted from Neural Network model has less error than other methods.

One of the important factors for economic development is the existence of an effective tax system. Higher tax revenues are also important to lower the aid dependency of low-income countries. Tax system can influence international investment decisions. They also encourage good governance and they are integral to strengthening state building and promote state accountability. The first step to understand public revenue system is to establish some commonly agreed performance measurement and benchmarks. One of the performance assessment benchmarks of tax system is tax capacity. In this regard in this study we investigated the determinant factors of tax capacity in oil exporter developing countries during 1996-2014 with “Bayesian Model Averaging” method to determine role and importance of any factors in affecting on tax capacity. The study intends to present country tax capacity typologies in order to fulfill emerging needs from developing countries’ policy makers.

Aim: The aim of the study is to examine the determinants of high-tech exports in developing countries during the period 1996 to 2013. One of the most important challenges of the empirical modeling is the selection of the potential variables that can be included in the econometric model, especially when there is a very wide range of explanatory variables. There is no acceptable way to solve this problem in the conventional econometric models. This article tries to fill this gap using Bayesian Model averaging and WALS econometrics approaches.
In this article we studied the determinants of export for 24 developing countries during the period 1996 to 2013 based on Bayesian Model Averaging (BMA) and Weighted-Average Least Square (WALS) technique.
Findings: The institutional quality (with an average coefficient of 2.12), human capital (0.85), the ratio of imports to GDP (0.04) and logarithm of GDP (1.87) have definite effects on the exports with inclusion probability of 100%. The sign of these coefficients are as expected. The human capital implies the creation of the endogenous knowledge. The knowledge-based capital is one of the key inputs for productive activities according to the endogenous growth theories so that more human capital improves the quality and productivity and ultimately leads to greater exports. A country which has a greater GDP, can supply more various products. Therefore, the trade greatly depends on the size of the country in terms of GDP. In fact, economies with higher income are more interested in specialization and sophistication of the products and have more trades. Import as an important channel of international knowledge spillover has a positive effect on the high-tech export. Import also affects the host country's export through various channels. In the long term, the countries can accelerate export sophistication through the dissemination of the foreign knowledge. Imports of intermediate and capital goods causes the transfer of the new technology into the country and reduce their production costs leading to high-tech exports. Some developing countries import large quantities of intermediate and tech-intensive goods and export them after simple assembling and processing the complex final products. The existence or establishment of appropriate institutions can empower the endogenous growth and productivity and therefore provide competitiveness and sustainable growth in a country's exports. Especially, improving the institutional quality and the rule of law increases the security of property rights and contracts enforcements and therefore creates a safe environment for the development of new markets, strengthening human capital, domestic research and development and information and communication technology. It also increases the rate of return of the capitals and the incentives for domestic and foreign investments through reducing the risk, and ultimately fortifies the competitiveness and exports. The ratio of capital to labor as well as research and development expenditure to GDP with the probability of 99% have negative effects on the high-tech exports. The weighted average of these two variable's coefficients are respectively -0.68 and -0.89. The increase of the ratio of capital to labor and research and development expenditure in the developing countries (under study) with poor production structures do not have expected positive effect on increasing the high-tech exports. If the necessary institutional infrastructures are not provided, the development of the physical capital and even the research and development expenditures encloses the country's natural resource-based and traditional industries. Other explanatory variables including the land area per capita, the real effective exchange rate, the population, the ratio of FDI to GDP and inflation due to the low inclusion probability do not affect the exports.
The traditional trade theories lack essential potentials to explain the tech-intensive exports in the developing countries. However, the institutions, the efficient human capital, higher GDP, the openness and the easier access to the foreign knowledge and investment can explain the behavior of the tech-intensive exports in the developing countries, confirming the main hypothesis of this article. Of course good governance maybe not perfect proxy of Institutional quality that should be addressed in future researches.