Autoregressive Method Research Articles

Structural Decomposition of the Impact of World Oil Price Shock Economic Growth in Indonesia

The aim of this research is to determine the response of economic growth in Indonesia to shocks or variable shocks in world oil prices, gold prices, inflation, exchange rates, and interest rates. This research uses time series data for 12 years (2010- 2021) using the Vector Auto Regressive (VAR) method. The results show that based on the results of the Impulse Response Function (IRF) test, world oil price shocks responded positively to economic growth in the 2nd period, 5-6th period, and 9-10th period, while the response was negative in the 3rd-4th period and the 2nd period. 7th-8th. Gold price shocks were responded positively by economic growth in periods 3-4 and 7-8, while the response was negative in periods 2, 5-6, and 9-10. Inflation shocks were responded to positively by economic growth in periods 2, 4-5, and 8-9, while negative responses were in periods 3, 6-7, and 10. Exchange rate shocks were responded to positively by economic growth in periods 3- 4 and 7-8, while the response was negative in the 2, 5-6, and 9-10 periods. Interest rate shocks responded positively to economic growth in periods 2, 5-6, and 9, while negative responses in periods 3-4, 7-8, and 10. Based on the results of the Variance Decomposition test, oil prices contributed 4,886%, gold contributed 5,545%, inflation contributed 0.306%, interest rates contributed 1,197%, and the exchange rate was the variable that contributed the most, namely 6,396%.

A spatiotemporal autoregressive neural network interpolation method for discrete environmental factors

The spatiotemporal interpolation model is necessary for generating continuous distributions for spatiotemporally discrete sampling points. However, there remain challenges in spatiotemporal interpolation due to the complex spatiotemporal effect and the imprecise kernel functions. Here, we proposed a spatiotemporal autoregressive neural network interpolation model (STARNN) that incorporates adaptive spatiotemporal distance quantification and supervised learning. The 10-fold cross-validation modelling on sea surface temperature and coastal nutrients demonstrated that the STARNN model performs better than baseline models and can well depict reasonable spatiotemporal distributions for environmental factors. By proposing two stacked neural networks, the STARNN model can accurately integrate spatial and temporal distances and avoids subjective selection of the kernel function. This study developed a novel interpolation model for processing discrete spatiotemporal points by following the data-driven paradigm, which can offer decision support for simulating the spread of sea temperature anomalies and optimizing the distribution of water quality measurement stations.

Exploring currency interdependence in West Africa: a time-varying parameter vector autoregression analysis

PurposeConsidering the impact of significant economic and political events, this study investigates the return spillovers and connectedness among eight West African currencies from March 31, 2010, to March 28, 2024. It aims to enhance understanding of the interdependencies within the West African foreign exchange market, providing insights into the region’s risk management and diversification opportunities.Design/methodology/approachUsing the time-varying parameter vector autoregression (TVP-VAR) method, this study analyzes daily exchange rate returns to capture the dynamic spillover effects and connectedness among the selected currencies. This approach identifies key transmitters and receivers of return shocks, reflecting the evolving interactions among the currencies over time.FindingsThe results show that the Sierra Leonean Leone, Cape Verdean Escudo, and West African CFA Franc are significant net transmitters of return shocks. At the same time, the Ghana Cedi, Nigerian Naira, Gambian Dalasi, Guinean Franc, and Liberian Dollar are net receivers, with the Gambian Dalasi being the most affected. These findings suggest relatively low regional spillover connectedness, offering favorable diversification opportunities.Originality/valueThis study provides a comprehensive analysis of the interconnectedness of West African currencies, contributing to the limited literature on this region. The findings have practical implications for investors and policymakers in managing foreign exchange risks and designing interventions to stabilize the market.

Safe Haven Assets In Stock Investments On The Indonesian Stock Exchange: An Empirical Study During The Covid-19 Pandemic

Safe haven assets are secure investment during times of economic downturn. This study aims at analysing the roles of three safe haven assets, Bitcoin, gold, and US dollars, during the COVID-19 pandemic. By conducting a time series analysis of returns on stocks, gold, Bitcoin, and US dollars between 2020 and 2022 using Vector Auto Regression (VAR) method, this study found the empirical evidence that suggests a negative correlation between gold and IDX, and between US dollars and IDX. Therefore, it can be conluded that gold and US dollars are safe haven assets in the IDX market. Meanwhile, despite being less resilient than gold and US dollars, Bitcoin’s indepence from stocks during bearish markets makes it a safe haven asset.

Monetary and Fiscal Policy on Inflation in Indonesia: Threshold Vector Autoregression Approach

The combination of monetary and fiscal policies is used to stabilize prices in Indonesia. The purpose of this study is to analyse the role of monetary and fiscal policies in stabilizing prices in Indonesia through the threshold concept. The data used is time series (2013Q1-2023Q2) with the Threshold Vector Autoregression method. The results show that the interest rate and budget thresholds can affect inflation stability. Other variables, such as the exchange rate, output gap, and FFR, have varying effects depending on the threshold. Thus, combining adjustments in monetary and fiscal policies and understanding the impact of inflation at various policymakers can better manage economic growth and maintain price stability in the long term.

State estimation for wind turbine systems

Wind Energy Systems is an area of research which is maturing rapidly due to the concern of long term global energy consumption, currently systems and dynamics research is ongoing in modeling and identification primarily for controller design, state estimation, fault detection and diagnosis. This paper presents a method to estimate and identify a wind turbine mathematical model, which includes structural deformation (the growth of wind turbine size, the loads wind turbine…) and aerodynamic effects. Finding a mathematical model that would be a good representation of the wind turbine system and at the same time suitable for controller design is a very difficult task for wind turbine system is strongly nonlinear. To get an estimate and an identification since of the mathematical model, we used a model detection algorithm, namely the autoregressive method and eigensystem realization algorithm. To improve the state prediction, we used optimal filtering by using the Kalman filter.

Comparative Study between AR-MEM Model and Decision Tree

Predictive modeling is understanding complex sys- tems and decisionmaking informed by the parameter space of various domains. The aim of this study is to compare two predictive models, the Auto-Regressive Maximum Entropy Model (ARMEM) and the Decision Tree model, for predicting the performance of a specific variable, Surface Ocean Direction, from a dataset. The dataset, obtained through High-Frequency Radar (HFR) measurements around Koko Head, was taken as a case study to test these models. ARMEM is a time-series model that merges the autoregressive methods with the principle of maximum entropy, thus being highly appropriate for high-resolution spectral analysis and noisy or incomplete data. On the other hand, the Decision Tree model works through recursive partitioning of data and thereby provides intuitive, interpretable predictions through capturing the underlying linear and nonlinear relationships.

The Effects of Macroprudential Policies on the Performance of Conventional Banks in Indonesia

Abstract A resilient banking system in Indonesia is essential to withstand economic fluctuations that have significantly impacted Indonesia, especially during financial crises. To address these challenges, effective macroprudential policies are required to support the development of a high-performance banking sector. This research examines the impact of macroprudential policies on the performance of conventional banks in Indonesia. The research uses time-series data from 2014 to 2023 obtained from the Indonesian Financial Services Authority (OJK) and the Central Bank of Indonesia (BI). The data is analysed using the Vector Autoregression (VAR) method. The variables estimated include macroprudential policy instruments, including the Macroprudential Intermediation Ratio, Countercyclical Capital Buffer, Loan-to-Value ratio, Minimum Statutory Reserves, and Return on Assets. The results of the study found that most macroprudential measures do not have a substantial impact on the performance of conventional banks in Indonesia. Only Minimum Statutory Reserves significantly affect Return on Assets. Additionally, most variables do not exhibit reciprocal relationships. However, some variables display unidirectional effects. Specifically, the Countercyclical Capital Buffer has a significant causal impact on Return on Assets, while Minimum Statutory Reserves also play a notable role in affecting Return on Assets. Moreover, there is a causal relationship between the Macroprudential Intermediation Ratio and Minimum Statutory Reserves, as well as between the Countercyclical Capital Buffer and Minimum Statutory Reserves. The results of this study can contribute to policies in helping regulators formulate strategies to deal with future economic crises by ensuring that existing policies can maintain financial stability.

Predicting cropland and fertilizer consumption models and their effect on crop production in interior Jiangsu Province: a distributed autoregressive lag method

Predicting cropland and fertilizer consumption models and their effect on crop production in interior Jiangsu Province: a distributed autoregressive lag method

Analysis of Outdoor and Indoor Radon Concentration Time Series Recorded with RadonEye Monitors

Consumer-grade economical radon monitors are becoming increasingly popular in private and institutional use, in the contexts of both Citizen Science and traditional research. Although originally designed for screening indoor radon levels in view of radon regulation and decisions about mitigation or remediation—motivated by the health hazard posed by high radon concentrations—researchers are increasingly exploring their potential in some environmental studies. For long time, radon has been used as a tracer for investigating atmospheric transport processes. This paper focuses on RadonEye, currently the most sensitive among low-cost monitors available on the market, and specifically, its potential use for monitoring very low radon concentrations. It has two objectives: firstly, discussing issues of statistics of low count rates, and secondly, analyzing radon concentration time series acquired with RadonEyes outdoors and in low-radon indoor spaces. Regarding the first objective, among other things, the inference radon concentration reported to expected true is discussed. The second objective includes the application of autoregressive methods and fractal statistics to time series analysis. The overall result is that radon dynamics can be well captured using this “low-tech” approach. Statistical results are plausible; however, few results are available in the literature for comparison, particularly concerning fractal methods. The paper may therefore be seen as an incentive for further research in this direction.

The Relationship Between Financial Development and Logistics Performance Index: A Panel ARDL Approach in Emerging Markets

The term "emerging markets" refers to countries that are undergoing economic and financial development. As financial development progresses in these nations, the capital available for international trade and the expansion of logistics infrastructure grows as well. This increased capital enables the financing of international trade activities and facilitates the execution of necessary logistics operations efficiently and effectively. Additionally, with enhanced financial development, it becomes easier to secure the capital required to improve logistics infrastructure and make it more functional. In this study, analyses are conducted using the Panel autoregressive distributed lag method in emerging market countries to evaluate the direct and indirect effects of financial development on logistics performance. The results reveal that increased access to financial institutions and markets positively affects logistics performance in both the short and long term. Among financial variables, the depth of financial institutions exhibits the strongest long-term influence, suggesting that a stable financial system is crucial for logistics efficiency. These insights provide valuable guidance for decision-makers aiming to strengthen logistics infrastructure and trade capabilities. By recognizing the role of financial development in supporting logistics performance, policymakers can craft strategies that support sustainable economic growth through improved logistics processes.

Real-Time Postural Disturbance Detection Through Sensor Fusion of EEG and Motion Data Using Machine Learning.

Millions of people around the globe are impacted by falls annually, making it a significant public health concern. Falls are particularly challenging to detect in real time, as they often occur suddenly and with little warning, highlighting the need for innovative detection methods. This study aimed to assist in the advancement of an accurate and efficient fall detection system using electroencephalogram (EEG) data to recognize the reaction to a postural disturbance. We employed a state-space-based system identification approach to extract features from EEG signals indicative of reactions to postural perturbations and compared its performance with those of traditional autoregressive (AR) and Shannon entropy (SE) methods. Using EEG epochs starting from 80 ms after the onset of the event yielded improved performance compared with epochs that started from the onset. The classifier trained on the EEG data achieved promising results, with a sensitivity of up to 90.9%, a specificity of up to 97.3%, and an accuracy of up to 95.2%. Additionally, a real-time algorithm was developed to integrate the EEG and accelerometer data, which enabled accurate fall detection in under 400 ms and achieved an over 99% accuracy in detecting unexpected falls. This research highlights the potential of using EEG data in conjunction with other sensors for developing more accurate and efficient fall detection systems, which can improve the safety and quality of life for elderly adults and other vulnerable individuals.

ANALISIS DAMPAK HARGA DAN VOLUME PERDAGANGAN TERHADAP RETURN SAHAM PERUSAHAAN SEKTOR PARIWISATA DI INDONESIA PERIODE SEBELUM DAN SELAMA COVID-19

Indonesia has many tourist destinations visited by local and foreign residents. So, Indonesia has great prospects for the tourism sector. However, Covid-19 caused tourism performance to decline as a result of government policy (PPKM). Covid-19 as an investment risk can undermine investor confidence in tourism sector stocks so that stock returns will be affected. This research aims to compare the influence of price and trading volume on tourism company stock returns before and during Covid-19 in Indonesia. This research uses the Vector Auto Regression method. The data used is a time series with a monthly time period from 2017 to 2022. This research found that the similarity of results in the period before and during Covid-19 shows that share prices are an important factor for low, medium and high return companies so that companies need to maintain market prices to be balanced. Low return companies are more sensitive to market shocks so they must be careful about risks. Trading volume is important for high return companies so it needs to be increased so that transactions also increase.

Sustainable improvement and application of multilingual english translation quality using T5 and MAML

In the current era of global economic integration and digital economy development, multilingual English translation plays a crucial role in cultural exchange. Traditional translation models have poor adaptability and fitting ability. To improve the translation effect and quality, this article combines the Text-To-Text Transfer Transformer (T5) and Model-Agnostic Meta-Learning (MAML) to study their sustainable improvement and application in multilingual English translation. In this article, the autoregressive learning method is first used to fine tune the pre-trained parameters of T5 model, and a generative multilingual English translation model is constructed. Then, combined with the MAML framework, the model is trained on multiple tasks to achieve rapid adaptation on new task data. Finally, a multilingual parallel corpus is constructed using web crawlers, and the translation model based on T5 and MAML is evaluated for quality using Bilingual Evaluation Understudy (BLEU) and Translation Error Rate (TER) as indicators. The experimental results show that compared to the baseline models Open Neural Machine Translation (OpenNMT), Transformer, and Open Parallel Corpus-Machine Translation (Opus-MT), the mean BLEU Score of the model in this article is 6.05%, 2.59%, and 2.05% higher, respectively. The conclusion indicates that the T5-MAML model can effectively improve the quality of multilingual English translation and achieve more natural and smooth translation output.

Uneven Burdens: The Intersection of Brownfields, Pollution, and Socioeconomic Disparities in New Jersey, USA

Brownfields, abandoned or underutilized lands having the potential for contamination due to industrial activities, pose health risks to nearby residents and raise environmental justice (EJ) concerns. This study examines the relationship between brownfield distribution and socioeconomic factors in New Jersey, a state with over 700 brownfield sites, using correlation, spatial autoregression, and instrumental variable regression methods. Results show that brownfields are associated with higher combined EJ stressors, indicating compounding environmental pollution and health risks in affected areas. Brownfields are disproportionately correlated with lower household income, higher rates of uninsured individuals, lower educational attainment, and a greater percentage of people of color. When accounting for spatial autocorrelation, relationships between income, education, and brownfield distributions are reduced, reflecting New Jersey’s historical, environmental, and social contexts. Overall, the presence of people of color maintains the strongest association with brownfield distribution. This study highlights ongoing EJ challenges faced by disadvantaged groups and emphasizes the need for targeted mitigation measures to enhance sustainability and equity. New Jersey serves as a critical case study given its ethnic diversity, growing urban population, and status as the state with the highest population density in the United States, illustrating broader trends of disproportionate exposure to environmental pollution among marginalized communities.

CoATR: A Convolutional Autoregressive Tensor-Ring decomposition method for sparse spatio-temporal traffic data

Spatio-temporal traffic data collected by various sensing systems have chronic issues of missing and corruption, thus accurate data imputation and prediction have been extensively researched. Previous most methods can be divided into two categories: model-driven methods with physical explanations and data-driven deep learning methods. These methods all achieve promising results due to their respective advantages. However they have some limitations: model-driven methods are often linear and may not accurately model the spatio-temporal complexity, while deep learning methods often lack the physical reality making them probe to over-fitting. Inspired by both, we propose a new Convolutional-based generalized Autoregressive Tensor-Ring decomposition method (CoATR) for the completion of spatio-temporal data. CoATR not only retains the advantages of the tensor-ring (TR) decomposition model for global modeling of spatio-temporal data but also exploits the ability of deep networks to model nonlinear features. To be specific, we introduce TR decomposition to capture the global low-rankness and employ multilayer convolutional neural networks to model the global complex interactions among the TR factors for exploring the nonlinear features of the spatio-temporal data. Moreover, we design a new autoregressive network to further explore the local temporal variation in the data. Extensive experiments on a variety of common traffic datasets have validated the effectiveness and superiority of the CoATR over classical model-driven methods and other state-of-the-art data-driven deep learning methods.

TCM: An efficient lightweight MLP-based network with affine transformation for long-term time series forecasting

Time series forecasting (TSF) involves extracting underlying patterns from past information to predict future sequences over a specific period. Extending the prediction length of time series and improving the prediction accuracy have always been challenging tasks. Autoregressive prediction methods based on Markov chains tend to accumulate errors over time. Although Transformer-based methods with various self-attention mechanisms have shown some improvements, they require higher memory and computational resources. In this work, we present an effective MLP-based TSF framework named TCM, which models the sequence and channel dependencies separately using Token MLP and Channel MLP. Additionally, we employ the Affine Transformation to replace layer normalization or batch normalization, leading to substantial enhancements in both accuracy and inference speed. Compared to current state-of-the-art long-term time series forecasting models, TCM achieves 6.0% relative improvement on seven real-world datasets, including electricity, weather, and illness domains. The TCM model, characterized by its efficiency and lightweight architecture, also makes it suitable for scenarios with high real-time requirements.

Inverse heat transfer for real-time thermal evaluation of aircraft thermal protection structure with embedded FBG sensors

Thermal Protection Structures (TPSs) are indispensable parts of aircraft. In the context of high-speed aviation vehicles, surface thermal information of the TPS is vital to ensuring the safe operation of the aircraft. However, the real-time measurement of surface thermal information becomes a great challenge due to the complex and volatile flight environment. In this paper, we propose a method for real-time monitoring of the TPS’ surface temperature by using embedded temperature sensors processed with inverse heat conduction model. In the investigation, tube packaged FBG (TP-FBG) sensors that were insensitive to strain and vibration were employed as embedded sensors in thermal protection composites. In the signal processing, thermal hysteresis correction model based on the Fourier heat conduction theory was established. After that, the data processing method of Auto-Regression with eXtra inputs was used to process the data of the sensors. The measurement error and response delay of bare FBG sensor, thermocouple and TP-FBG sensors are compared by a transient thermal shock experiment. The experiment results show that thermal hysteresis effect of the TP-FBG sensors is significant. However, the correction model can effectively correct the measurement errors of TP-FBG caused by thermal hysteresis. Building upon the model, two inverse heat conduction methods with adaptive boundaries were proposed. These methods can retrieve surface thermal information by processing the measured temperatures from embedded sensors. Numerical simulations demonstrate that while the instantaneous heating moment may exhibit significant relative errors, the average relative error remains below 5.54%. Experimental validation using a quartz lamp to heat real TPS materials further confirms the effectiveness of the inversion method. The method successfully mitigates the influence of process deviations during sensor embedding and material preparation, maintaining an average relative error of less than 8.53%. These findings suggest the potential for the proposed correction model and inversion methods in practical engineering applications. Particularly, their applicability to dynamic high temperature monitoring of thermal protection composites.

Impact of oil and gold prices on Bitcoin price during Russia-Ukraine and Israel-Gaza wars

Political tensions between countries and geopolitical events can create significant uncertainty in financial markets, impacting the prices of traditional assets such as oil and gold, as well as the emerging asset, Bitcoin. This study investigates the effect of gold and oil prices, including their price shocks, on Bitcoin price during two specific conflicts: the Russia-Ukraine war and the Israel-Gaza conflict. To achieve this, we employed the Vector Auto-Regression (VAR) method and wavelet coherence analysis. The data used spans daily observations from January 2022 to April 2024.The findings indicate that gold prices have a positive impact on Bitcoin prices. Specifically, there is a predominantly short-term relationship between gold prices and Bitcoin prices throughout the entire period under review, with this influence gradually decreasing over the medium and long term. Conversely, the relationship between Brent oil prices and Bitcoin prices during the same period shows that crude oil prices generally have a negative and insignificant effect on Bitcoin prices in the short term.

Deep Learning and Handcrafted Features for Thyroid Nodule Classification

ABSTRACTIn this research, we present a refined image‐based computer‐aided diagnosis (CAD) system for thyroid cancer detection using ultrasound imagery. This system integrates a specialized convolutional neural network (CNN) architecture designed to address the unique aspects of thyroid image datasets. Additionally, it incorporates a novel statistical model that utilizes a two‐dimensional random coefficient autoregressive (2D‐RCA) method to precisely analyze the textural characteristics of thyroid images, thereby capturing essential texture‐related information. The classification framework relies on a composite feature vector that combines deep learning features from the CNN and handcrafted features from the 2D‐RCA model, processed through a support vector machine (SVM) algorithm. Our evaluation methodology is structured in three phases: initial assessment of the 2D‐RCA features, analysis of the CNN‐derived features, and a final evaluation of their combined effect on classification performance. Comparative analyses with well‐known networks such as VGG16, VGG19, ResNet50, and AlexNet highlight the superior performance of our approach. The outcomes indicate a significant enhancement in diagnostic accuracy, achieving a classification accuracy of 97.2%, a sensitivity of 84.42%, and a specificity of 95.23%. These results not only demonstrate a notable advancement in the classification of thyroid nodules but also establish a new standard in the efficiency of CAD systems.