Forecasting the future prevalence of inflammatory bowel disease in Korea through 2048: an epidemiologic study employing autoregressive integrated moving average models

BACKGROUND AND AIM

The global inflammatory bowel disease (IBD) escalation has precipitated an increased disease burden and economic impact, particularly in Asia. This study primarily aimed to predict the future prevalence of IBD in Korea and elucidate its evolution pattern.

METHODS

Using a validated diagnostic algorithm, we analyzed data from the Korean National Health Insurance Service between 2004 and 2017 to identify patients with IBD. We predicted the number and prevalence of patients with IBD from 2018 to 2048 with the autoregressive integrated moving average method. A generalized linear model (GLM) was also employed to identify factors contributing to the observed trend in IBD prevalence.

RESULTS

Our prediction model validation demonstrated an acceptable error range for IBD prevalence, with a 2.45% error rate and a mean absolute difference of 2.61. We foresee a sustained average annual increase of 4.51 IBD cases per 100 000, culminating in a prevalence of 239.73 per 100 000 by 2048. The forecasted average annual percent change was 6.17% for males and 2.75% for females over the next 30 years. The GLM analysis revealed that age, gender and time significantly impact the prevalence of IBD, with notable disparities observed between genders in specific age groups for both Crohn's disease and ulcerative colitis (all interaction P < 0.05).

CONCLUSIONS

Our study forecasts a notable increase in Korean IBD prevalence by 2048, particularly among males and the 20-39 age group, highlighting the need to focus on these high-risk groups to mitigate the future disease burden.

Prophet time series modeling of waste disposal rates in four North American cities

In this study, three different univariate municipal solid waste (MSW) disposal rate forecast models (SARIMA, Holt-Winters, Prophet) were examined using different testing periods in four North American cities with different socioeconomic conditions. A review of the literature suggests that the selected models are able to handle seasonality in a time series; however, their ability to handle outliers is not well understood. The Prophet model generally outperformed the Holt-Winters model and the SARIMA model. The MAPE and R2 of the Prophet model during pre-COVID-19 were 4.3-22.2% and 0.71-0.93, respectively. All three models showed satisfactory predictive results, especially during the pre-COVID-19 testing period. COVID-19 lockdowns and the associated regulatory measures appear to have affected MSW disposal behaviors, and all the univariate models failed to fully capture the abrupt changes in waste disposal behaviors. Modeling errors were largely attributed to data noise in seasonality and the unprecedented event of COVID-19 lockdowns. Overall, the modeling errors of the Prophet model were evenly distributed, with minimum modeling biases. The Prophet model also appeared to be versatile and successfully captured MSW disposal rates from 3000 to 39,000 tons/month. The study highlights the potential benefits of the use of univariate models in waste forecast.

Time series models in prediction of severe fever with thrombocytopenia syndrome cases in Shandong province, China

Severe fever with thrombocytopenia syndrome (SFTS) is an emerging infectious disease caused by the SFTS virus (SFTSV). Predicting the incidence of this disease in advance is crucial for policymakers to develop prevention and control strategies. In this study, we utilized historical incidence data of SFTS (2013-2020) in Shandong Province, China to establish three univariate prediction models based on two time-series forecasting algorithms Autoregressive Integrated Moving Average (ARIMA) and Prophet, as well as a special type of recurrent neural network Long Short-Term Memory (LSTM) algorithm. We then evaluated and compared the performance of these models. All three models demonstrated good predictive capabilities for SFTS cases, with the predicted results closely aligning with the actual cases. Among the models, the LSTM model exhibited the best fitting and prediction performance. It achieved the lowest values for mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE). The number of SFTS cases in the subsequent 5 years in this area were also generated using this model. The LSTM model, being simple and practical, provides valuable information and data for assessing the potential risk of SFTS in advance. This information is crucial for the development of early warning systems and the formulation of effective prevention and control measures for SFTS.

Impacts of crude oil market on global economy: Evidence from the Ukraine-Russia conflict via fuzzy models

The increasing Russia-Ukraine crisis is without a doubt Europe's most prominent conflict since World War II, changing the dynamics of the oil and other key markets. Because the oil market has traditionally interacted with other financial and commodity markets, it will be intriguing to examine how it interacts with substantial financial assets amid market volatility induced by a conflict. The goal of this study is to propose a fuzzy time series (FTS) model and to compare its competitiveness to existing fuzzy time series (FTS) models, Autoregressive Integrated Moving Average (ARIMA) model and some machine learning methods i.e. Artificial Neural Networks (ANN), Support Vector Machine (SVM) and XGBoost models. We considered changes in the partitioning universe of discourse, optimization of parameters method(s), and interval estimation to make the forecast accuracy more precise forecasting than traditional methods via MAPE. The event-based data results show the proposed fuzzy time series model is outperforming all the competitive methods in the study. Furthermore, the proposed model forecasting shows a future decline tendency in WTi market crude oil prices (US$/BBL) after being at the record highest level, which is good news for the worldwide economy.

Global, regional and national trends in tuberculosis incidence and main risk factors: a study using data from 2000 to 2021

BACKGROUND

Despite the significant progress over the years, Tuberculosis remains a major public health concern and a danger to global health. This study aimed to analyze the spatial and temporal characteristics of the incidence of tuberculosis and its risk factors and to predict future trends in the incidence of Tuberculosis.

METHODS

This study used secondary data on tuberculosis incidence and tuberculosis risk factor data from 209 countries and regions worldwide between 2000 and 2021 for analysis. Specifically, this study analyses the spatial autocorrelation of Tuberculosis incidence from 2000 to 2021 by calculating Moran's I and identified risk factors for Tuberculosis incidence by multiple stepwise linear regression analysis. We also used the Autoregressive Integrated Moving Average model to predict the trend of Tuberculosis incidence to 2030. This study used ArcGIS Pro, Geoda and R studio 4.2.2 for analysis.

RESULTS

The study found the global incidence of Tuberculosis and its spatial autocorrelation trends from 2000 to 2021 showed a general downward trend, but its spatial autocorrelation trends remained significant (Moran's I = 0.465, P < 0.001). The risk factors for Tuberculosis incidence are also geographically specific. Low literacy rate was identified as the most pervasive and profound risk factor for Tuberculosis.

CONCLUSIONS

This study shows the global spatial and temporal status of Tuberculosis incidence and risk factors. Although the incidence of Tuberculosis and Moran's Index of Tuberculosis are both declining, there are still differences in Tuberculosis risk factors across countries and regions. Even though literacy rate is the leading risk factor affecting the largest number of countries and regions, there are still many countries and regions where gender (male) is the leading risk factor. In addition, at the current rate of decline in Tuberculosis incidence, the World Health Organization's goal of ending the Tuberculosis pandemic by 2030 will be difficult to achieve. Targeted preventive interventions, such as health education and regular screening of Tuberculosis-prone populations are needed if we are to achieve the goal. The results of this study will help policymakers to identify high-risk groups based on differences in TB risk factors in different areas, rationalize the allocation of healthcare resources, and provide timely health education, so as to formulate more effective Tuberculosis prevention and control policies.

A bibliometric literature review of stock price forecasting: From statistical model to deep learning approach

We introduce a comprehensive analysis of several approaches used in stock price forecasting, including statistical, machine learning, and deep learning models. The advantages and limitations of these models are discussed to provide an insight into stock price forecasting. Traditional statistical methods, such as the autoregressive integrated moving average and its variants, are recognized for their efficiency, but they also have some limitations in addressing non-linear problems and providing long-term forecasts. Machine learning approaches, including algorithms such as artificial neural networks and random forests, are praised for their ability to grasp non-linear information without depending on stochastic data or economic theory. Moreover, deep learning approaches, such as convolutional neural networks and recurrent neural networks, can deal with complex patterns in stock prices. Additionally, this study further investigates hybrid models, combining various approaches to explore their strengths and counterbalance individual weaknesses, thereby enhancing predictive accuracy. By presenting a detailed review of various studies and methods, this study illuminates the direction of stock price forecasting and highlights potential approaches for further studies refining the stock price forecasting models.

Forecasting the Spread of COVID-19 Using Deep Learning and Big Data Analytics Methods

To contain the spread of the COVID-19 pandemic, there is a need for cutting-edge approaches that make use of existing technology capabilities. Forecasting its spread in a single or multiple countries ahead of time is a common strategy in most research. There is, however, a need for all-inclusive studies that capitalize on the entire regions on the African continent. This study closes this gap by conducting a wide-ranging investigation and analysis to forecast COVID-19 cases and identify the most critical countries in terms of the COVID-19 pandemic in all five major African regions. The proposed approach leveraged both statistical and deep learning models that included the autoregressive integrated moving average (ARIMA) model with a seasonal perspective, the long-term memory (LSTM), and Prophet models. In this approach, the forecasting problem was considered as a univariate time series problem using confirmed cumulative COVID-19 cases. The model performance was evaluated using seven performance metrics that included the mean-squared error, root mean-square error, mean absolute percentage error, symmetric mean absolute percentage error, peak signal-to-noise ratio, normalized root mean-square error, and the R2 score. The best-performing model was selected and used to make future predictions for the next 61 days. In this study, the long short-term memory model performed the best. Mali, Angola, Egypt, Somalia, and Gabon from the Western, Southern, Northern, Eastern, and Central African regions, with an expected increase of 22.77%, 18.97%, 11.83%, 10.72%, and 2.81%, respectively, were the most vulnerable countries with the highest expected increase in the number of cumulative positive cases.

Impact of global value chain and technological innovation on China's industrial greenhouse gas emissions and trend prediction

The global value chain has introduced profound changes in international trade, economic development, and technology progress as well as greenhouse gas emissions worldwide. This paper investigated the impact of the global value chain and technological innovation on greenhouse gas emissions by introducing a partially linear functional-coefficient model based on panel data of 15 industrial sectors in China from 2000 to 2020. Moreover, the greenhouse gas emission trends of China's industrial sectors from 2024 to 2035 were predicted using the autoregressive integrated moving average model. The results showed that (1) Greenhouse gas emissions were affected negatively by global value chain position and independent innovation. Nevertheless, foreign innovation had the opposite effect. (2) The results of the partially linear functional-coefficient model implied that the inhibitory effect of independent innovation on GHG emissions decreased with an improvement in the global value chain position. (3) The positive effect of foreign innovation on greenhouse gas emissions increased and then, decreased as the global value chain position improved. (4) The prediction results indicated that greenhouse gas emissions will continue on an upward trend from 2024 to 2035, while industrial carbon dioxide emissions should peak at 10.21 Gt in 2028. This carbon-peaking goal would be achieved in China's industrial sector by actively improving the global value chain position. Addressing these issues will enable China to take full advantage of the development opportunities of participating in the global value chain.

A forecasting model based on ARIMA and artificial neural networks for end-OF-life vehicles

The accelerated growth of the automotive supply network has had an immeasurable impact on the environment, especially relating to reusing and disposal of materials. The appropriate management of End-of-Life Vehicles (ELV) has become an imperative item for achieving sustainable development in the field of interest and it is, therefore, a target of special attention from global economies in recent years. Therefore, the present study aims to estimate the future generation of ELVs to assist decision making and mitigate the global impact of this type of waste on the environment. For this, a hybrid forecasting model was used, based on Autoregressive Integrated Moving Average (ARIMA) methodology and on Artificial Neural Networks (ANN), with a set of temporal data extracted from Brazilian sectoral platforms. The results achieved point to a good convergence of the model, indicating better performance than a naive or trivial prediction. The efficiency obtained by the Nash-Sutcliffe coefficient was 98% and the expectation is that for the year 2030, approximately 5.2 million ELVs will be produced in Brazil, of which only 78 thousand units would be effectively recycled, considering the current vehicle recycling rate in the country. Considering the scarcity of information that supports decision-making in waste management in Brazil, this study may also contribute to the proposition of alternatives that favor the proper management of automotive waste, providing a reference for the formulation and implementation of policies related to ELVs in the country.

State of the art in total pulse production in major states of India using ARIMA techniques

Pulses are staple protein-rich food for Indian vegetarians, and India is one of the largest producers in the world. The present investigation is an attempt to study the trend in the production of total pulses in India using the autoregressive integrated moving average (ARIMA) method. For stochastic trend estimation, yearly data were used for the period from 1961 to 2019. On the basis of the performance of several goodness of model fit criteria, the most suitable ARIMA model is chosen to capture the trend of pulse production. Forecasting for the 10 years from 2020 to 2029 is done, and it is observed that India has the highest forecast value (31.03302 million tonnes) in 2029. This study will play an important role in determining the gap between production of and demand for pulses in the future.

Using LSTM and ARIMA to Simulate and Predict Limestone Price Variations

There have been many improvements and advancements in the application of neural networks in the mining industry. In this study, two advanced deep learning neural networks called recurrent neural network (RNN) and autoregressive integrated moving average (ARIMA) were implemented in the simulation and prediction of limestone price variation. The RNN uses long short-term memory layers (LSTM), dropout regularization, activation functions, mean square error (MSE), and the Adam optimizer to simulate the predictions. The LSTM stores previous data over time and uses it in simulating future prices based on defined parameters and algorithms. The ARIMA model is a statistical method that captures different time series based on the level, trend, and seasonality of the data. The auto ARIMA function searches for the best parameters that fit the model. Different layers and parameters are added to the model to simulate the price prediction. The performance of both network models is remarkable in terms of trend variability and factors affecting limestone price. The ARIMA model has an accuracy of 95.7% while RNN has an accuracy of 91.8%. This shows that the ARIMA model outperforms the RNN model. In addition, the time required to train the ARIMA is than that of the RNN. Predicting limestone prices may help both investors and industries in making economical and technical decisions, for example, when to invest, buy, sell, increase, and decrease production.

Time series prediction of under-five mortality rates for Nigeria: comparative analysis of artificial neural networks, Holt-Winters exponential smoothing and autoregressive integrated moving average models

BACKGROUND

Accurate forecasting model for under-five mortality rate (U5MR) is essential for policy actions and planning. While studies have used traditional time series modeling techniques (e.g., autoregressive integrated moving average (ARIMA) and Holt-Winters smoothing exponential methods), their appropriateness to predict noisy and non-linear data (such as childhood mortality) has been debated. The objective of this study was to model long-term U5MR with group method of data handling (GMDH)-type artificial neural network (ANN), and compare the forecasts with the commonly used conventional statistical methods-ARIMA regression and Holt-Winters exponential smoothing models.

METHODS

The historical dataset of annual U5MR in Nigeria from 1964 to 2017 was obtained from the official website of World Bank. The optimal models for each forecasting methods were used for forecasting mortality rates to 2030 (ending of Sustainable Development Goal era). The predictive performances of the three methods were evaluated, based on root mean squared errors (RMSE), root mean absolute error (RMAE) and modified Nash-Sutcliffe efficiency (NSE) coefficient. Statistically significant differences in loss function between forecasts of GMDH-type ANN model compared to each of the ARIMA and Holt-Winters models were assessed with Diebold-Mariano (DM) test and Deming regression.

RESULTS

The modified NSE coefficient was slightly lower for Holt-Winters methods (96.7%), compared to GMDH-type ANN (99.8%) and ARIMA (99.6%). The RMSE of GMDH-type ANN (0.09) was lower than ARIMA (0.23) and Holt-Winters (2.87). Similarly, RMAE was lowest for GMDH-type ANN (0.25), compared with ARIMA (0.41) and Holt-Winters (1.20). From the DM test, the mean absolute error (MAE) was significantly lower for GMDH-type ANN, compared with ARIMA (difference = 0.11, p-value = 0.0003), and Holt-Winters model (difference = 0.62, p-value< 0.001). Based on the intercepts from Deming regression, the predictions from GMDH-type ANN were more accurate (β₀ = 0.004 ± standard error: 0.06; 95% confidence interval: - 0.113 to 0.122).

CONCLUSIONS

GMDH-type neural network performed better in predicting and forecasting of under-five mortality rates for Nigeria, compared to the ARIMA and Holt-Winters models. Therefore, GMDH-type ANN might be more suitable for data with non-linear or unknown distribution, such as childhood mortality. GMDH-type ANN increases forecasting accuracy of childhood mortalities in order to inform policy actions in Nigeria.

[Application of autoregressive integrated moving average model in predicting the reported notifiable communicable diseases in China]

Objective: To develop the models for predicting the reported legally notifiable diseases in China. Autoregressive integrated moving average (ARIMA) model was applied to forecast the trend of diseases. Methods: Cases used for building the model were from of the records of Notifiable Infectious Diseases in China from May 2009 to July 2016 with R software and the model's predictive ability was tested by the data from August 2016 to January 2017. Results: A strong seasonal nature was seen in the reported cases of notifiable communicable diseases, with the lowest point in February and highest peak in June. ARIMA (4, 1, 0) (1, 1, 1)(12) model was established by the team to forecast the notifiable communicable diseases. Data showed that the biggest and lowest relative errors appeared as 9.78% and 2.21%, respectively, with the mean of the relative error as 5.39%. Conclusion: Based on the results of this study, the ARIMA (4, 1, 0) (1, 1, 1)(12) model seemed to have had the sound prediction of notifiable communicable diseases in China.

Global trends in the awareness of sepsis: insights from search engine data between 2012 and 2017

BACKGROUND

Sepsis is an established global health priority with high mortality that can be curtailed through early recognition and intervention; as such, efforts to raise awareness are potentially impactful and increasingly common. We sought to characterize trends in the awareness of sepsis by examining temporal, geographic, and other changes in search engine utilization for sepsis information-seeking online.

METHODS

Using time series analyses and mixed descriptive methods, we retrospectively analyzed publicly available global usage data reported by Google Trends (Google, Palo Alto, CA, USA) concerning web searches for the topic of sepsis between 24 June 2012 and 24 June 2017. Google Trends reports aggregated and de-identified usage data for its search products, including interest over time, interest by region, and details concerning the popularity of related queries where applicable. Outlying epochs of search activity were identified using autoregressive integrated moving average modeling with transfer functions. We then identified awareness campaigns and news media coverage that correlated with epochs of significantly heightened search activity.

RESULTS

A second-order autoregressive model with transfer functions was specified following preliminary outlier analysis. Nineteen significant outlying epochs above the modeled baseline were identified in the final analysis that correlated with 14 awareness and news media events. Our model demonstrated that the baseline level of search activity increased in a nonlinear fashion. A recurrent cyclic increase in search volume beginning in 2012 was observed that correlates with World Sepsis Day. Numerous other awareness and media events were correlated with outlying epochs. The average worldwide search volume for sepsis was less than that of influenza, myocardial infarction, and stroke.

CONCLUSIONS

Analyzing aggregate search engine utilization data has promise as a mechanism to measure the impact of awareness efforts. Heightened information-seeking about sepsis occurs in close proximity to awareness events and relevant news media coverage. Future work should focus on validating this approach in other contexts and comparing its results to traditional methods of awareness campaign evaluation.

The impact of alcohol policies on alcohol-attributable diseases in Taiwan-A population-based study

BACKGROUND

Taiwan has some of the strictest alcohol-related driving laws in the world. However, its laws continue to be toughened to reduce the ever-increasing social cost of alcohol-related harm.

AIM

This study assumes that alcohol-related driving laws show a spillover effect such that behavioral changes originally meant to apply behind the wheel come to affect drinking behavior in other contexts. The effects of alcohol driving laws and taxes on alcohol-related morbidity are assessed; incidence rates of alcohol-attributable diseases (AAD) serve as our measure of morbidity.

METHODS

Monthly incidence rates of alcohol-attributable diseases were calculated with data from the National Health Insurance Research Database (NHIRD) from 1996 to 2011. These rates were then submitted to intervention analyses using Seasonal Autoregressive Integrated Moving Average models (ARIMA) with multivariate adaptive regression splines (MARS). ARIMA is well-suited to time series analysis while MARS helps fit the regression model to the cubic curvature form of the irregular AAD incidence rates of hospitalization (AIRH).

RESULTS

Alcoholic liver disease, alcohol abuse and dependence syndrome, and alcohol psychoses were the most common AADs in Taiwan. Compared to women, men had a higher incidence of AADs and their AIRH were more responsive to changes in the laws governing permissible blood alcohol. The adoption of tougher blood alcohol content (BAC) laws had significant effects on AADs, controlling for overall consumption of alcoholic beverages.

CONCLUSION

Blood alcohol level laws and alcohol taxation effectively reduced alcohol-attributable morbidities with the exception of alcohol dependence and abuse, a disease to which middle-aged, lower income people are particularly susceptible. Attention should be focused on this cohort to protect this vulnerable population.

Potential of combined ultrasound and microneedles for enhanced transdermal drug permeation: a review

Transdermal drug delivery (TDD) is limited by the outer layer of the skin, i.e., the stratum corneum. Research on TDD has become very active in the recent years and various technologies have been developed to overcome the resistance of the stratum corneum to molecular diffusion. In particular, researchers have started to consider the possibility of combining the TDD technologies in order to have further increase in drug permeability. Both microneedles (MNs) and ultrasound are promising technologies. They achieve enhancement in drug permeation via different mechanisms and therefore give a good potential for combining with each other. This review will focus on discussing the potential of this combinational technique along with other important issues, e.g., the mechanisms of ultrasound and MNs as it is and these mechanisms which are coupled via the two systems (i.e. MNs and ultrasound). We discuss the possible ways to achieve this combination as well as how this combination would increase the permeability. Some of the undeveloped (weaker) research areas of MNs and sonophoresis are also discussed in order to understand the true potential of combining the two technologies when they are developed further in the future. We propose several hypothetical combinations based on the possible mechanisms involved in MNs and ultrasound. Furthermore, we carry out a cluster analysis by which we determine the significance of this combinational method in comparison with some other selected combinational methods for TDD (e.g., MNs and iontophoresis). Using a time series analysis tool (ARIMA model), the current trend and the future development of combined MNs and ultrasound are also analysed. Overall, the review in this paper indicates that combining MNs and ultrasound is a promising TDD method for the future.

Time series analysis of contaminant transport in the subsurface: applications to conservative tracer and engineered nanomaterials

Accurately predicting the transport of contaminants in the field is subject to multiple sources of uncertainty due to the variability of geological settings, the complexity of field measurements, and the scarcity of data. Such uncertainties can be amplified when modeling some emerging contaminants, such as engineered nanomaterials, when a fundamental understanding of their fate and transport is lacking. Typical field work includes collecting concentration at a certain location for an extended period of time, or measuring the movement of plume for an extended period time, which would result in a time series of observation data. This work presents an effort to evaluate the possibility of applying time series analysis, particularly, autoregressive integrated moving average (ARIMA) models, to forecast contaminant transport and distribution in the subsurface environment. ARIMA modeling was first assessed in terms of its capability to forecast tracer transport at two field sites, which had different levels of heterogeneity. After that, this study evaluated the applicability of ARIMA modeling to predict the transport of engineered nanomaterials at field sites, including field measured data of nanoscale zero valent iron and (nZVI) and numerically generated data for the transport of nano-fullerene aggregates (nC60). This proof-of-concept effort demonstrates the possibility of applying ARIMA to predict the contaminant transport in the subsurface environment. Like many other statistical models, ARIMA modeling is only descriptive and not explanatory. The limitation and the challenge associated with applying ARIMA modeling to contaminant transport in the subsurface are also discussed.