Autoregressive Method Research Articles

Unveiling health dynamics: Exploring the impact of CO2 emissions, urbanization, and renewable energy on life expectancy and infant mortality in SAARC countries (1990–2022)

AbstractThis study examines the intensifying threat to public health caused by increased CO2 emissions from energy and modern urbanization. In this connection, we focused on SAARC countries from 1990 to 2022. The study explores the dynamic association among urbanization, CO2 emissions, renewable energy, and key health indicators, namely life expectancy and infant mortality. It offers nuanced insights by highlighting the requirement for sustainable policies to meet health challenges associated with CO2 emissions and urban development in SAARC economies. To achieve the study objectives, the authors utilized panel auto regressive distributed lag for assessing short‐term and long‐run effects, and the method of moments quantile regressions to check these effects across different quantiles. The empirical results underscore the positive impact of renewable energy, urbanization, GDP, and industrialization on life expectancy. Conversely, CO2 emissions exhibit a detrimental impact on life expectancy, leading towards numerous diseases in both the short and long term. Notably, in the case of infant mortality, the study discloses that renewable energy, urbanization, GDP, and industrialization negatively affect infant mortality, while CO2 emissions positively impact infant mortality in both short and long‐run scenarios. Fully modified ordinary least squares and dynamic ordinary least squares further fortified our findings, supporting the results derived from panel auto regressive distributed lag and method of moments quantile regressions. The study's policy implications highlight the imperative for governments and policymakers to prioritize renewable energy and sustainable urbanization, mitigating the adverse health effects of CO2 emissions from non‐renewable energy sources. The study's findings further endorse a strategic shift towards renewable energy sources, positioning them as substitutes for conventional forms such as fossil fuels. Additionally, the emphasis is on planned and sustainable urbanization, facilitating improved access to health facilities and overall public health.

DiffSeqMol: A Non-Autoregressive Diffusion-Based Approach for Molecular Sequence Generation and Optimization

Background: The application of deep generative models for molecular discovery has witnessed a significant surge in recent years. Currently, the field of molecular generation and molecular optimization is predominantly governed by autoregressive models regardless of how molecular data is represented. However, an emerging paradigm in the generation domain is diffusion models, which treat data non-autoregressively and have achieved significant breakthroughs in areas such as image generation. Methods: The potential and capability of diffusion models in molecular generation and optimization tasks remain largely unexplored. In order to investigate the potential applicability of diffusion models in the domain of molecular exploration, we proposed DiffSeqMol, a molecular sequence generation model, underpinned by diffusion process. Results & Discussion: DiffSeqMol distinguishes itself from traditional autoregressive methods by its capacity to draw samples from random noise and direct generating the entire molecule. Through experiment evaluations, we demonstrated that DiffSeqMol can achieve, even surpass, the performance of established state-of-the-art models on unconditional generation tasks and molecular optimization tasks. Conclusion: Taken together, our results show that DiffSeqMol can be considered a promising molecular generation method. It opens new pathways to traverse the expansive chemical space and to discover novel molecules.

Relationship between fintech by Google search and bank stock return: a case study of Vietnam

Due to the ongoing global debate regarding the relationship between fintech and banks, including developing countries, this study aims to investigate this relationship in the case of Vietnam, an emerging nation. The study analyzes the relationship between fintech search and bank stock returns, which are measures of fintech and banks, respectively. The time series data for fintech and bank stock returns were obtained from Google Trends and Vietstock, respectively. Exploratory factor analysis was utilized to derive the fintech variables, while the bank stock return variable was calculated using a basket of eight listed banks from 2017w46 to 2021w46. The results were estimated using the vector autoregression and Granger causality method and validated with the copula method. A key finding of this study is the presence of a simultaneous negative change and bidirectional causality between bank stock returns and fintech lending. Furthermore, several other interesting findings were discovered: (1) the causal relationship from fintech to bank stock returns is weaker compared with the opposite direction; (2) unidirectional causality exists between different types of fintech, such as influence from FinFintech to FinLending, from FinPayment to FinLending and FinWallet, from FinMoney to FinFintech, from FinWallet to FinLending, and from FinProduct to FinFintech; and (3) there is an equal occurrence of simultaneous increase or decrease between bank stock returns and certain types of fintech, specifically between BankReturn and FinPayment, BankReturn and FinLending, as well as BankReturn and FinWallet. These findings shed light on the complex relationship between fintech and banks, offering insights that contribute to our understanding of this dynamic interplay in the context of Vietnam’s emerging fintech landscape.

Text-Guided Molecule Generation with Diffusion Language Model

Text-guided molecule generation is a task where molecules are generated to match specific textual descriptions. Recently, most existing SMILES-based molecule generation methods rely on an autoregressive architecture. In this work, we propose the Text-Guided Molecule Generation with Diffusion Language Model (TGM-DLM), a novel approach that leverages diffusion models to address the limitations of autoregressive methods. TGM-DLM updates token embeddings within the SMILES string collectively and iteratively, using a two-phase diffusion generation process. The first phase optimizes embeddings from random noise, guided by the text description, while the second phase corrects invalid SMILES strings to form valid molecular representations. We demonstrate that TGM-DLM outperforms MolT5-Base, an autoregressive model, without the need for additional data resources. Our findings underscore the remarkable effectiveness of TGM-DLM in generating coherent and precise molecules with specific properties, opening new avenues in drug discovery and related scientific domains. Code will be released at: https://github.com/Deno-V/tgm-dlm.

Assessing the long-term impact of macroeconomic and environment dynamics: Does sustainable energy production shape the environmental landscape of south ASIAN nations?

<p class="MsoNormal" style="margin-top: 10pt; text-align: justify;"><span lang="EN-US" style="font-family: 'times new roman', times, serif; font-size: 14pt;">This study delves into the intricate interconnections among the financial sector, macroeconomic dynamics, and the advancement of sustainable energy production in South Asia. Utilizing data from 1987 to 2022 in Bangladesh, India, Pakistan, and Sri Lanka, the research employs the Structural Vector Autoregressive (SVAR) method to analyze financial market responses. The findings reveal nuanced insights into the complex relationship between economic and environmental factors, fostering a comprehensive understanding of the region's intricate nexus between financial markets and sustainable energy initiatives. In the Economic Growth model, the negative coefficient for CO2 emissions underscores environmental concerns, indicating an inverse relationship with economic growth. Conversely, positive coefficients for non-fossil energy production, Gross Fixed Capital Formation, stock exchange value, and domestic credit emphasize their positive impact on economic growth. A positive GDP coefficient in the CO2 emission model signifies heightened carbon emissions with economic growth, while a negative coefficient for non-fossil energy production suggests potential environmental benefits. The nearly unitary positive coefficient for Gross Fixed Capital Formation highlights its significant role in driving carbon emissions. The Non-Fossil Energy Production model reveals an inverse relationship between GDP and clean energy, suggesting a potential reliance on fossil fuels during economic growth. Positive coefficients for CO2 emissions, Gross Fixed Capital Formation, stock exchange value, and domestic credit highlight factors influencing non-fossil energy production. Long-run stable coefficients affirm variable significance, while a negative error correction term suggests convergence in developing countries. These models offer nuanced insights into the intricate relationships among economic, environmental, and energy variables, guiding sustainable development policies in South Asia.</span></p>

On autoregressive and neural methods for massive-MIMO channel de-noising

In modern wireless communication systems, the Multiple-Input Multiple-Output (MIMO) technology allows to greatly increase power efficiency, the serving area, and the overall cell throughput through the use of the antenna array beamforming. Nevertheless, the MIMO systems require accurate channel state knowledge to apply correct precoding. In 5G Time Division Duplex (TDD) systems, the Channel State Information (CSI) is obtained via Sounding Reference Signals (SRS) transmitted by the User Equipment (UE). UEs have limited power capabilities and thus cannot achieve high Uplink (UL) Signal-to-Noise Ratio (SNR) on gNodeB (gNB) in large bandwidth. There are multiple techniques that can be applied to improve the accuracy of Channel Estimation (CE) in noisy conditions. In this paper, we describe a classical method, namely the Vector Autoregression (VAR) with adaptive model order estimation, as well as a modern Deep Neural Network (DNN) approach for the massive-MIMO channel estimation de-noising problem. The developed methods and signal pre and postprocessing steps are described, followed by their performance evaluation in a set of realistic simulations. The designed algorithms provide results outperforming the baseline spatio-temporal windowing approaches by approximately equal to 2dB effective Downlink (DL) Signal-to-Interference-plus-Noise Ratio (SINR) metric in single and multi-user MIMO scenarios. Extensive simulation results demonstrate the robustness of the developed methods to the dynamic channel conditions.

Modelling the determinants of birth tourism demand: a time series analysis

ABSTRACT Birth tourism has been steadily rising in recent years as potential mothers travel across borders with the main purpose of giving birth. However, the economic reasons behind this form of tourism have not been adequately examined in the existing literature. This study explores the determinants of birth tourism by incorporating factors such as the average income of visitors, price level, exchange rate, and infant mortality while accounting for the potential impact of the Citizenship Amendment Act 2005 in New Zealand. The variables have been introduced into the model using a standard tourism demand framework and a new dynamic autoregressive distributed lag method has been used in the study. Ultimately, rising income of visitors and exchange rate are reported to boost birth tourism while rising price reduces birth tourism. The result also shows that the implementation of the Citizenship Amendment Act 2005 has no significant effect on birth tourism, which implies that monetary and financial considerations such as the income of tourists, price level, and exchange are the most important drivers of birth tourism in New Zealand.

APPLICATION OF THE NEURAL NETWORK AUTOREGRESSIVE (NNAR) METHOD FOR FORECASTING THE VALUE OF OIL AND GAS EXPORTS IN INDONESIA

Indonesia is one of the countries with the most diversity and abundant natural resources, consisting of many commodities, and has enormous trade potential with other countries The success of economic activity a country can be measured by the amount of economic growth that occurs in the country. A recession is when a country's economic condition is getting worse. Meanwhile, a recession in Indonesia is expected to occur in 2023. In a 2022 news issue written by the editorial team, tirto.id said that some experts say that if 2023 is a recession, the cause is due to a spike in inflation from the impact of the Russia-Ukraine conflict. It is known that the value of oil and gas exports affects the Indonesian economy. Any increase in the value of oil and gas exports will be followed by an increase in economic growth, and vice versa. However, over time, the value of oil and gas exports has decreased every year. Therefore, forecasting the value of oil and gas exports is needed so that the country's economic sector development strategy can be on target. In addition, oil and gas export forecasting is also needed to determine the distribution of goods exports that must be carried out. In this study, we forecast the value of oil and gas exports using the neural network autoregressive (NNAR) method. The choice of this method is made because there is no assumption of normality of the residuals and white noise like in autoregressive models. From the NNAR method, the best model results are obtained, namely NNAR (2,3) with a MAPE value of 11.75640%, which means that this model has very good forecasting performance.

China's risk contagion using the mixed-frequency macro-financial network

We explore and compare the fundamental laws of risk contagion under different shocks during the global financial crisis and COVID-19 pandemic periods. Using the daily returns of financial market indices and the monthly growth rates of macroeconomic data for two periods, from January 1, 2003, to December 31, 2010, and from January 1, 2015, to December 31, 2021, we construct a macro-financial network with the mixed-frequency vector autoregression (MF-VAR) method to produce well-defined measures of intra-system and cross-system risk contagion. We investigate the differences in risk contagion under different types of major events using an event study model. The empirical results show that the financial system is the center of risk contagion. Regarding the direction of the cross-system risk contagion, the risk contagion from the financial to the macroeconomic system dominates the contagion in the opposite direction. The impacts of financial and pandemic shocks on the macro-financial risk contagion differ significantly. Importantly, both investor sentiment and consumer confidence mediate these impacts, providing useful insights into the prevention of risk contagion.

Determinants of Economic Growth: A Granger Causality Analysis

ABSTRACT: Countries devote a lot of time and effort to develop economically, but they nevertheless struggle to attain that growth consistently. Research into economic growth should have provided some guidance as to which actions these countries should adopt, but multiple studies using different methods and datasets generally have led to an array of outcomes that contradict one another. However, Granger causality analysis can provide better insight into the relationship between some key variables and economic growth to give these countries more direction. Granger causality results establish a stronger connection between variables, even though they do not depict true causation. We therefore use a Vector Auto Regression method with Generalized Method of Moments estimation and then Granger-causality to discover how commonly used variables in the literature individually impact economic growth and in what way for a selection of OECD and non-OECD countries. We choose government expenditure, foreign direct investment (FDI), education (as a proxy for human capital), population growth, urbanization, and trade openness as our variables of interest. We also utilize updated datasets from the World Development Indicators (WDI) ranging from the years 2000-2016, based on the availability of the data. We find that FDI and population growth do Granger-cause economic development for OECD countries but that government expenditures, education, urbanization, and trade openness do not show such Granger causality. However economic growth Granger-causes government expenditures, FDI, and education for these OECD countries. For the non-OECD countries, we find that FDI, population growth, and trade openness Granger-cause growth but Granger causality does not occur from government expenditure, education, and urbanization to economic development. Yet similar to the OECD countries, economic growth does Granger-cause FDI and trade openness for the non-OECD countries as well. The overall results indicate that all countries should encourage the expansion of foreign direct investment to grow their economy, which then in turn can lead to additional investment as well as more resources available for the country as a whole. We also suggest further studies should focus on the economic development of individual countries rather than regions, particularly in terms of how technology can drive growth and impact these other variables.

An Autoregressive Transformer Model For Crisis Management Using Twitter Tweets

A disaster is a major event that lasts for a long time and causes a lot of damage to people, property, the economy, or the environment. The damage is so bad that the group or society that is affected can't handle it on its own. When a disaster happens, it costs developing countries the most. More than 95% of all deaths from hazards happen in developing countries, and as a share of gross domestic product, losses from natural hazards are 20 times higher in developing countries than in developed nations. To minimize the loss at the time of disaster and to overcome it quickly, a survey is taken on how to manage disaster effectively using Machine Learning and social networks. Social media platforms like Twitter, Instagram and Facebook acts as a communication medium during disaster and also help to take preventive measures before the occurrence of the disaster. Machine learning algorithms also helps in response and recovery phase after the occurrence of the disaster. The paper aims to provides an elaborated review on the concepts of Machine Learning (ML) along with social media and addresses how various other technologies like use of physical sensors, Remote sensors, Graphical Information System (GIS), Internet of Things (IoT) and Neural Network[1]s like Artificial Neural Network (ANN), Deep Neural Network (DNN) can combine with Machine Learning algorithms for the management of disaster efficiently using social media. Finally, based on the study, a generalized autoregressive pre-training method called XLNet is proposed which overcomes the limitations of BERT using its autoregressive formulation and directions for the future research is explained further.

Is labour force participation independent of unemployment? A panel analysis for high-income countries

PurposeWe test the pertinence of the unemployment invariance hypothesis (UIH) for a set of Organisation for Economic Co-operation and Development countries.Design/methodology/approachWe empirically investigate the nexus between unemployment and labour force participation employing structural vector autoregressive methods for panel data.FindingsWe find that shocks in unemployment produce long-lasting, negative effects on participation, testifying to a discouraged worker effect.Originality/valueOur results do not support the validity of the UIH in high-income economies. This has relevant implications for policy making and macroeconomic models.

The relationship between monetary stability and central bank independence: The case of Azerbaijan

The independence of the central bank is one of the most important factors for effective monetary policy. Central bank independence is closely related to monetary stability, which is an important part of monetary policy. In this study, the purpose of the analysis is to understand whether monetary stability functions effectively for central bank independence in Azerbaijan using the vector autoregression method. In addition, the Granger Causality test was conducted to empirically investigate how central bank independence affects the provision of monetary stability in the economy of Azerbaijan over the data period from 1996 to 2022. In this framework, indices or variables are the exchange rate stability index (ERS) in 1996–2022, the level of monetary independence index (MI) in 1996–2020, taken from the “trilemma indexes”, which are defined as the consumer price index (CPI) in 1996–2022, and the broad money supply (M2) in 1996–2022. The findings of the study show that the independence of the central bank has a positive effect on the monetary stability of the Azerbaijani economy.

Hybrid convolutional long short‐term memory models for sales forecasting in retail

AbstractThis study proposes novel sales forecasting approaches that merge deep learning methods in a hybrid model. Long short‐term memory (LSTM) is adopted for modeling the temporal characteristics of the data, whereas the convolutional neural network (CNN) focuses on identifying and extracting relevant exogenous information. We propose stacked (S‐CNN‐LSTM) and parallel (P‐CNN‐LSTM) hybrid architectures to understand complex time series data with varying seasonal patterns and multiple products correlations. The performance drivers of both architectures were empirically tested with a real‐world multivariate retail dataset and outperformed when compared with simple neural network architectures and standard autoregressive methods for short and long‐term forecasting horizons. When compared with traditional predictive approaches, the proposed hybrid models reduce the computational complexity while providing flexibility and robustness.

Bootstrapping Long-Run Covariance of Stationary Functional Time Series

A key summary statistic in a stationary functional time series is the long-run covariance function that measures serial dependence. It can be consistently estimated via a kernel sandwich estimator, which is the core of dynamic functional principal component regression for forecasting functional time series. To measure the uncertainty of the long-run covariance estimation, we consider sieve and functional autoregressive (FAR) bootstrap methods to generate pseudo-functional time series and study variability associated with the long-run covariance. The sieve bootstrap method is nonparametric (i.e., model-free), while the FAR bootstrap method is semi-parametric. The sieve bootstrap method relies on functional principal component analysis to decompose a functional time series into a set of estimated functional principal components and their associated scores. The scores can be bootstrapped via a vector autoregressive representation. The bootstrapped functional time series are obtained by multiplying the bootstrapped scores by the estimated functional principal components. The FAR bootstrap method relies on the FAR of order 1 to model the conditional mean of a functional time series, while residual functions can be bootstrapped via independent and identically distributed resampling. Through a series of Monte Carlo simulations, we evaluate and compare the finite-sample accuracy between the sieve and FAR bootstrap methods for quantifying the estimation uncertainty of the long-run covariance of a stationary functional time series.

Faktor Penentu Pertumbuhan Ekonomi Indonesia

This research aims to determine: the effect of inflation on Indonesia's economic growth, the effect of HDI on Indonesia's economic growth and the effect of unemployment on economic growth in Indonesia. The data used in this research is secondary data sourced from the websites of various agencies including BPS, aseanstats.org, and kidb.adb.org, from 1993-2022. This research uses the Autoregressive Distributed Lag or ARDL analysis tool model. The ARDL model is a combination of autoregressive (AR) and distributed lag (DL) methods. The AR method is a method that uses one or more past data from the dependent variable, while DL is a regression method that involves data at the present and past times from the independent variable. This model can differentiate short-term and long-term responses of the variables studied. The findings of this study indicate that there is cointegration between the variables used in this study so that the ARDL test can be carried out. The results of short-term data processing show that inflation has had a negative and significant effect on Indonesia's economic growth in the current year, two and three years previously. Meanwhile, HDI has a significant influence in the current year and the previous two years. Meanwhile, unemployment has had a significant negative influence in the current year on Indonesia's economic growth. In long-term estimates, inflation and unemployment have a negative and significant effect on economic growth, however, HDI does not have a significant effect on Indonesia's economic growth.

Combination of autoregressive graphical models and time series bootstrap methods for risk management in marine insurance

In this paper a methodology to assess risk by forecasting the trend of marine losses at a global scale is presented. The proposed procedure, which can be used to continuously update an insurance company’s costing model, identifies the most relevant risk indicators through Probabilistic Graphical Models (PGMs). The use of PGMs makes the variable selection more understandable since they provide a clear interface to interpret the model and perform predictions. Furthermore, this procedure can be used to verify independence relationships, validate the dataset and identify unexpected links among the considered variables. The robustness of estimates, crucial for risk assessment in the insurance context, is dealt with bootstrap.

Indonesia's FDI in Development of Special Economic Zones (SEZs)

In realizing equitable development and economic growth, it is necessary to develop special economic zones (SEZs) as a form of commitment of the Indonesian government; and if viewed from the budget perspective applied, it is still limited. This study aims to determine the factors affecting Foreign Direct Investment (FDI) in Indonesia from 2008 2nd quarter to 2021 2nd quarter. Based on that period, the economy in Indonesia experienced a contraction since the Asian monetary crisis in 1998. So that the development of SEZs experienced financial budget constraints and Indonesia can try to find other financing solutions. For developing countries, such initiatives should be taken in order to attract foreign investment. This study uses quantitative research with the analytical method used is the Vector Error Correction Model (VECM), which is a derivative of the vector autoregressive (VAR) method. The results of this study indicate that for the long-term VECM variable estimates only interest rates have a negative effect on FDI, and trade inhibition, inflation, and GDP variables have a positive effect. While the wage variable has no effect on FDI. In this phenomenon, it is necessary to pay attention first to the influence on FDI, companies in Indonesia, the majority of which are labor-intensive workers.
 Keywords: FDI, GDP, inflation, interest rates, trade inhibition, wage

Fragility or habitualized corporate behavior? Corporate and macroeconomic determinants of debt dollarization: Evidence from Turkey

AbstractThis study examines the firm‐level and macroeconomic‐level determinants of debt dollarization, which is a critical vulnerability for a key emerging market: Turkey. The study examines the firm‐level and macroeconomic determinants of debt dollarization between 2005 and 2017 using the generalized method of moments and the panel vector autoregressive method, both of which are highly innovative. The results show that manufacturing firms tend to dollarize debt, while macroeconomic variables such as the real exchange rate, inflation, and credit expansion significantly affect debt dollarization. Moreover, debt dollarization was found to be a habit of manufacturing firms in the Turkish economy.

ANALISIS PENGARUH VARIABEL EKONOMI MAKRO TERHADAP NILAI TUKAR RUPIAH

Exchange rate is a key part of a country’s economic stability. The US Federal Reserve has been aggressively hiked the federal fund rates (Fed Rate), which potentially affect Indonesia’s rupiah exchange rate. Hence, it is interesting to understand the effect of Fed Rate hikes alongside other exchange rate determinants, i.e., BI Rate, inflation, trade balance, and foreign exchange reserves on the value of Rupiah exchange rate. This study uses Vector Autoregression (VAR) method to analyze data between 2018 to 2023. In line with what has often been assumed, this study found that Fed Rate hikes contribute to depreciation the value of Rupiah. On the other hand, policy reinforcement to maintain other macroeconomic indicators, such as inflation, trade balance position, and foreign exchange reserves will support the value of rupiah. Thus, it is essential for the government and the central bank of Indonesia to continuously monitor global economic policies and give adequate and immediate response to maintain Rupiah value.