The COVID-19 pandemic has been an extraordinary event, the type of event that rarely occurs but that has major impacts on the stock market. The pandemic has created high volatility and caused extreme fluctuations in the stock market. The stock market can be characterized as either linear or nonlinear. One method that can detect extreme fluctuations is extreme value theory (EVT). This study employed a semisystematic literature review on the use of the EVT method to estimate investment risk in the stock market. The literature used was selected by applying the preferred reporting items for systematic review and meta-analyses (PRISMA) guidelines, sourced from the ScienceDirect.com, ProQuest, and Scopus databases. A bibliometric analysis was conducted to determine the study characteristics and identify any research gaps. The results of the analysis show that studies on this topic are rarely carried out. Research in this field is generally performed only in univariate cases and is very complicated in multivariate cases. Given these limitations, further research could focus on developing a conceptual model that is dynamic and sensitive to extreme fluctuations, with multivariable inputs, in order to predict investment risk. The model developed here considered the variables that affect stock price fluctuations as the input data. The combination of VaR–EVT and machine-learning methods is effective in increasing model accuracy because it combines linear and nonlinear models.

(1) Background: sentiment analysis is a computational technique employed to discern individuals opinions, attitudes, emotions, and intentions concerning a subject by analyzing reviews. Machine learning-based sentiment analysis methods, such as Support Vector Machine (SVM) classification, have proven effective in opinion classification. Feature selection methods have been employed to enhance model performance and efficiency, with the Chi-Square method being a commonly used technique; (2) Methods: this study analyzes user reviews of Indonesia’s National Health Insurance (Mobile JKN) application, evaluating model performance and identifying optimal hyperparameters using the F1-Score metric. Sentiment analysis is conducted using a combined approach of SVM classification and Chi-Square feature selection; (3) Results: the sentiment analysis of user reviews for the Mobile JKN application reveals a predominant tendency towards positive reviews. The best model performance is achieved with an F1-Score of 96.82%, employing hyperparameters where $C$ is set to 10 and a “linear” kernel; (4) Conclusions: this study highlights the effectiveness of SVM classification and the significance of Chi-Square feature selection in sentiment analysis. The findings offer valuable insights into users’ sentiments regarding the Mobile JKN application, contributing to the improvement of user experience and advancing the field of sentiment analysis.

This article presents a study on forecasting silver prices using the extreme gradient boosting (XGBoost) machine learning method with hyperparameter tuning. Silver, a valuable precious metal used in various industries and medicine, experiences significant price fluctuations. XGBoost, known for its computational efficiency and parallel processing capabilities, proves suitable for predicting silver prices. The research focuses on identifying optimal hyperparameter combinations to improve model performance. The study forecasts silver prices for the next six days, evaluating models based on mean absolute percentage error (MAPE) and root mean square error (RMSE). Model A (the best model based on MAPE value) suggests silver prices decline on the first and second days, rise on the third, decline again on the fourth, and stabilize with an increase on the fifth and sixth days. Model A achieves a MAPE of 5.98% and an RMSE of 1.6998, utilizing specific hyperparameters. Conversely, model B (the best model based on RMSE value) indicates a price decrease until the third day, followed by an upward trend until the sixth day. Model B achieves a MAPE of 6.06% and an RMSE of 1.6967, employing distinct hyperparameters. The study also compared the proposed models with several other ensemble models (CatBoost and random forest). The model comparison was carried out by incorporating 2 additional metrics (MAE and SI), and it was found that the proposed models exhibited the best performance. These findings provide valuable insights for forecasting silver prices using XGBoost.

Gold price fluctuations have a significant impact because gold is a haven asset. When financial markets are volatile, investors tend to turn to safer instruments such as gold, so gold price forecasting becomes important in economic uncertainty. The novelty of this research is the comparative analysis of time series forecasting models using ARIMA and the NNAR methods to predict gold price movements specifically applied to gold price data with non-stationary and non-linear characteristics. The aim is to identify the strengths and limitations of ARIMA and NNAR on such data. ARIMA can only be applied to time series data that are already stationary or have been converted to stationary form through differentiation. However, ARIMA may struggle to capture complex non-linear patterns in non-stationary data. Instead, NNAR can handle non-stationary data more effectively by modeling the complex non-linear relationships between input and output variables. In the NNAR model, the lag values of the time series are used as input variables for the neural network. The dataset used is the closing price of gold with 1449 periods from January 2, 2018, to October 5, 2023. The augmented Dickey-Fuller test dataset obtained a p-value = 0.6746, meaning the data is not stationary. The ARIMA(1, 1, 1) model was selected as the gold price forecasting model and outperformed other candidate ARIMA models based on parameter identification and model diagnosis tests. Model performance is evaluated based on the RMSE and MAE values. In this study, the ARIMA(1, 1, 1) model obtained RMSE = 16.20431 and MAE = 11.13958. The NNAR(1, 10) model produces RMSE = 16.10002 and MAE = 11.09360. Based on the RMSE and MAE values, the NNAR(1, 10) model produces better accuracy than the ARIMA(1, 1, 1) model.

The sustainable development goals (SDGs) provide an integrated framework of targets and indicators, including the elimination of stunting, to support better development planning. Indonesia faces a significant challenge as it ranks fourth globally in terms of stunting prevalence, exacerbated by disparities across regions, gender, and socioeconomic status, further compounded by the ongoing COVID-19 pandemic. Given the interlinked nature of SDGs, this study provides empirical support for the prioritization of SDG indicators, primarily in the context of stunting elimination at the district level in Indonesia. This study employed a combination of economic complexity and network theory, utilizing data from a comprehensive set of 54 indicators spanning 28 targets within 13 SDG goals in 514 districts. The analysis is based on network metrics, including revealed comparative advantage (RCA), proximity, centrality, and density to establish the SDG interlinkage network and identify key priority indicators. The findings highlight the importance of prioritizing indicators such as civil registration, health facilities and services, access to basic facilities and housing, and access to ICT in efforts to reduce stunting, particularly among disadvantaged households. Given the unique resources and capacities of each region, our analysis offers district-specific prioritization strategies for stunting elimination.