Ani Budi Astuti
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ani Budi Astuti has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Suci Astutik
Department of Statistics, Faculty of Mathematics and Natural Sciences, Universitas Brawijaya, Indonesia
3 shared publications - Adji Achmad Rinaldo Fernandes
Brawijaya University
2 shared publications - Fachira Haneinanda Junianto
Brawijaya University
2 shared publications - Muhammad Hisyam Lee
Department of Mathematical Sciences, Universiti Teknologi Malaysia, Johor Bahru 81310, Malaysia
2 shared publications - Rismania Hartanti Putri Yulianing Damayanti
Department Statistics, Universitas Brawijaya
2 shared publications - Solimun Solimun
Brawijaya University
2 shared publications
Latest Publications
- Open Access
- 30 November 2026
- Published by Maulana Malik Ibrahim State Islamic University in CAUCHY: Jurnal Matematika Murni dan Aplikasi
- Vol. 11 (2) , 2456-2474
Modified Ramsey RESET and Double-Resampling Bootstrap for Semiparametric Path Modeling of Financial Literacy
Financial literacy is a multidimensional construct that is becoming increasingly important in the digital age, yet modeling its determinants often poses challenges when some of the relationships among variables are nonlinear and the normality assumption is not met. This study compared the standard bootstrap and the double-resampling bootstrap within a semiparametric path analysis framework based on truncated splines, applied to the estimation of indirect (mediation) effects in financial literacy modeling. Primary data were collected from 100 college students in Malang City using purposive sampling, with technology access (X1) as the exogenous variable, financial knowledge (Y1) as the mediator, and financial literacy (Y2) as the response variable. The Ramsey RESET linearity test indicated that the relationship between technology access and financial knowledge was nonlinear, and it was therefore modeled using a one-knot linear truncated spline (minimum GCV of 0.155), while the other paths were linear. A Monte Carlo study at four sample sizes (n = 50, 100, 200, and 1000) with R = 400 replications per cell evaluated both procedures on coverage probability, interval length, test size, and power in addition to bias, RMSE, and relative efficiency. Two formulations of the double-resampling standard error are distinguished. Under the B2-scaled formulation, the formulation that reproduces the empirical results reported here, the relative efficiency equals the number of second-stage repetitions B2 exactly, at every coefficient and every sample size, which follows as an algebraic identity rather than as a property of the estimator. The resulting intervals were shorter by a factor of 1/B2 but attained an empirical coverage of 0.387 against a nominal 0.95, and the associated test rejected a true null hypothesis in 0.612 of replications against a nominal 0.05; neither quantity improved as n rose from 50 to 1000. Under the averaged-estimate formulation, that is, the definition as written, calibration was restored (coverage 0.950, test size 0.036) and the efficiency advantage was no longer present (relative efficiency 0.92). The standard bootstrap was well calibrated throughout (coverage 0.939, test size 0.046) and was therefore adopted as the basis for inference in this model. Technology access had a significant direct effect on financial literacy under both procedures; the indirect effect through financial knowledge was positive on both sides of the knot and was established at the 5% level under the standard bootstrap applied directly to the empirical estimates (tcalc = 4.523, p 0.001 below the knot; tcalc = 3.215, p = 0.002 above it), even though the calibrated procedure attained a simulated power of only 0.520 below the knot and 0.355 above it for effects of the estimated size at n = 100.
- 30 November 2026
- Published by Maulana Malik Ibrahim State Islamic University in CAUCHY: Jurnal Matematika Murni dan Aplikasi
- Vol. 11 (2) , 2456-2474
Open AccessModified Ramsey RESET and Double-Resampling Bootstrap for Semiparametric Path Modeling of Financial Literacy
Financial literacy is a multidimensional construct that is becoming increasingly important in the digital age, yet modeling its determinants often poses challenges when some of the relationships among variables are nonlinear and the normality assumption is not met. This study compared the standard bootstrap and the double-resampling bootstrap within a semiparametric path analysis framework based on truncated splines, applied to the estimation of indirect (mediation) effects in financial literacy modeling. Primary data were collected from 100 college students in Malang City using purposive sampling, with technology access (X1) as the exogenous variable, financial knowledge (Y1) as the mediator, and financial literacy (Y2) as the response variable. The Ramsey RESET linearity test indicated that the relationship between technology access and financial knowledge was nonlinear, and it was therefore modeled using a one-knot linear truncated spline (minimum GCV of 0.155), while the other paths were linear. A Monte Carlo study at four sample sizes (n = 50, 100, 200, and 1000) with R = 400 replications per cell evaluated both procedures on coverage probability, interval length, test size, and power in addition to bias, RMSE, and relative efficiency. Two formulations of the double-resampling standard error are distinguished. Under the B2-scaled formulation, the formulation that reproduces the empirical results reported here, the relative efficiency equals the number of second-stage repetitions B2 exactly, at every coefficient and every sample size, which follows as an algebraic identity rather than as a property of the estimator. The resulting intervals were shorter by a factor of 1/B2 but attained an empirical coverage of 0.387 against a nominal 0.95, and the associated test rejected a true null hypothesis in 0.612 of replications against a nominal 0.05; neither quantity improved as n rose from 50 to 1000. Under the averaged-estimate formulation, that is, the definition as written, calibration was restored (coverage 0.950, test size 0.036) and the efficiency advantage was no longer present (relative efficiency 0.92). The standard bootstrap was well calibrated throughout (coverage 0.939, test size 0.046) and was therefore adopted as the basis for inference in this model. Technology access had a significant direct effect on financial literacy under both procedures; the indirect effect through financial knowledge was positive on both sides of the knot and was established at the 5% level under the standard bootstrap applied directly to the empirical estimates (tcalc = 4.523, p 0.001 below the knot; tcalc = 3.215, p = 0.002 above it), even though the calibrated procedure attained a simulated power of only 0.520 below the knot and 0.355 above it for effects of the estimated size at n = 100.
- Open Access
- 30 November 2026
- Published by Maulana Malik Ibrahim State Islamic University in CAUCHY: Jurnal Matematika Murni dan Aplikasi
- Vol. 11 (2) , 1734-1753
Bayesian Framework for Error Correction Model-Nonlinear Autoregressive Distributed Lag
This study proposes a Bayesian framework for estimating the Error Correction Model–Nonlinear Autoregressive Distributed Lag (ECM–NARDL) to analyze the dynamic and asymmetric relationship between inflation and economic growth in Indonesia. The model is estimated using Gibbs Sampling within a Markov Chain Monte Carlo (MCMC) framework, allowing parameter uncertainty to be evaluated through posterior distributions. Annual data from 1990–2024 are used for empirical analysis. Unit root tests indicate that the variables are integrated of order one, while the Bounds test confirms the existence of a long-run equilibrium relationship. The estimation results reveal a significant error-correction mechanism, suggesting that deviations from long-run equilibrium are corrected relatively quickly. The findings also indicate asymmetric effects of inflation, where decreases in inflation support long-run economic growth, while short-run increases in inflation negatively affect growth. Bayesian diagnostics and posterior predictive checks confirm the stability and adequacy of the proposed Bayesian ECM–NARDL framework.
- 30 November 2026
- Published by Maulana Malik Ibrahim State Islamic University in CAUCHY: Jurnal Matematika Murni dan Aplikasi
- Vol. 11 (2) , 1734-1753
Open AccessBayesian Framework for Error Correction Model-Nonlinear Autoregressive Distributed Lag
This study proposes a Bayesian framework for estimating the Error Correction Model–Nonlinear Autoregressive Distributed Lag (ECM–NARDL) to analyze the dynamic and asymmetric relationship between inflation and economic growth in Indonesia. The model is estimated using Gibbs Sampling within a Markov Chain Monte Carlo (MCMC) framework, allowing parameter uncertainty to be evaluated through posterior distributions. Annual data from 1990–2024 are used for empirical analysis. Unit root tests indicate that the variables are integrated of order one, while the Bounds test confirms the existence of a long-run equilibrium relationship. The estimation results reveal a significant error-correction mechanism, suggesting that deviations from long-run equilibrium are corrected relatively quickly. The findings also indicate asymmetric effects of inflation, where decreases in inflation support long-run economic growth, while short-run increases in inflation negatively affect growth. Bayesian diagnostics and posterior predictive checks confirm the stability and adequacy of the proposed Bayesian ECM–NARDL framework.
- Open Access
- 20 October 2026
- Published by Pandawa Institute in Journal of Multidisciplinary Applied Natural Science
Bayesian and Non-Bayesian Semiparametric Path Models with Truncated Splines: Simulation and Case Study on Financial Literacy
The complex development of socioeconomic data requires an analytical approach that is capable of adequately capturing nonlinear structural relationships and inferential uncertainty. This study aims to compare Bayesian and non-Bayesian approaches in truncated spline-based semiparametric path analysis, both in terms of statistical inference and methodological implications, using financial literacy as a case study. The non-Bayesian approach is estimated using weighted least squares (WLS) with significance testing based on double bootstrap resampling, while the Bayesian approach is constructed through likelihood, prior, and posterior specifications with inference based on credible intervals. In addition to the empirical analysis, a simulation study evaluates the performance of both approaches across a range of sample sizes, noise levels, degrees of nonlinearity, knot specifications, and error structures. Within the simulation scenarios considered in this study, the Bayesian approach was more stable at small to medium sample sizes, whereas the non-Bayesian approach showed greater efficiency at large sample sizes. Methodologically, these findings confirm that the superiority of inferential approaches is contextual and highly influenced by data characteristics and analysis objectives. Rather than proposing a new estimator, this study contributes a systematic comparative framework in which Bayesian and non-Bayesian inference are applied to an identical truncated spline semiparametric path model and evaluated side by side. It also extends the application of truncated splines in structural models. The proposed framework is expected to serve as a methodological reference for socioeconomic research involving nonlinear relationships and indirect effects under conditions of data uncertainty.
- 20 October 2026
- Published by Pandawa Institute in Journal of Multidisciplinary Applied Natural Science
Open AccessBayesian and Non-Bayesian Semiparametric Path Models with Truncated Splines: Simulation and Case Study on Financial Literacy
The complex development of socioeconomic data requires an analytical approach that is capable of adequately capturing nonlinear structural relationships and inferential uncertainty. This study aims to compare Bayesian and non-Bayesian approaches in truncated spline-based semiparametric path analysis, both in terms of statistical inference and methodological implications, using financial literacy as a case study. The non-Bayesian approach is estimated using weighted least squares (WLS) with significance testing based on double bootstrap resampling, while the Bayesian approach is constructed through likelihood, prior, and posterior specifications with inference based on credible intervals. In addition to the empirical analysis, a simulation study evaluates the performance of both approaches across a range of sample sizes, noise levels, degrees of nonlinearity, knot specifications, and error structures. Within the simulation scenarios considered in this study, the Bayesian approach was more stable at small to medium sample sizes, whereas the non-Bayesian approach showed greater efficiency at large sample sizes. Methodologically, these findings confirm that the superiority of inferential approaches is contextual and highly influenced by data characteristics and analysis objectives. Rather than proposing a new estimator, this study contributes a systematic comparative framework in which Bayesian and non-Bayesian inference are applied to an identical truncated spline semiparametric path model and evaluated side by side. It also extends the application of truncated splines in structural models. The proposed framework is expected to serve as a methodological reference for socioeconomic research involving nonlinear relationships and indirect effects under conditions of data uncertainty.
- Open Access
- 30 September 2026
- Published by Universitas Airlangga in Jurnal Respirasi
- Vol. 12 (3) , 203-211
Integrated Exhaled Breath VOC Profiling and Serum sICAM-1 for Biomarker Characterization in Drug-Sensitive and Drug-Resistant Tuberculosis
Introduction: Rapid, non-invasive screening for drug-resistant tuberculosis (DR-TB) is urgently needed in high-burden settings such as Indonesia, where sputum-based diagnostics have practical limitations. Integrating exhaled breath volatile organic compounds (VOCs), which reflect host–pathogen metabolic interactions, with serum soluble intercellular adhesion molecule-1 (sICAM-1), a marker of endothelial activation and inflammation, may improve early detection. This study evaluated the combined potential of VOC profiling and sICAM-1 to distinguish drug-sensitive TB (DS-TB), DR-TB, and healthy controls. Methods: A prospective cross-sectional study was conducted in four TB referral hospitals in East Java, Indonesia. Eighty-eight participants were enrolled: DS-TB (n=32), DR-TB (n=30), and healthy controls (n=26). Exhaled breath VOC profiles were analyzed using a breath analyzer, and serum sICAM-1 levels were measured by enzyme-linked immunosorbent assay. Group comparisons used ANOVA or Kruskal–Wallis tests with post hoc analyses, and correlations were assessed using Pearson or Spearman tests. Results: Serum sICAM-1 levels were significantly higher in both TB groups than in controls (p<0.01) but did not differ between DS-TB and DR-TB (p=0.961). Several VOC components differed significantly, particularly formaldehyde, acetone, and methane (CH₄). CH₄ levels were significantly higher in DR-TB than in DS-TB and controls (p<0.05). A weak negative correlation between formaldehyde and sICAM-1 was observed in DS-TB (r = −0.355; p = 0.046). Conclusion: Integrating VOC profiling with sICAM-1 assessment offers a promising non-invasive strategy for TB screening. While sICAM-1 reflects inflammatory activity, methane-based VOC signatures may improve identification of DR-TB.
- 30 September 2026
- Published by Universitas Airlangga in Jurnal Respirasi
- Vol. 12 (3) , 203-211
Open AccessIntegrated Exhaled Breath VOC Profiling and Serum sICAM-1 for Biomarker Characterization in Drug-Sensitive and Drug-Resistant Tuberculosis
Introduction: Rapid, non-invasive screening for drug-resistant tuberculosis (DR-TB) is urgently needed in high-burden settings such as Indonesia, where sputum-based diagnostics have practical limitations. Integrating exhaled breath volatile organic compounds (VOCs), which reflect host–pathogen metabolic interactions, with serum soluble intercellular adhesion molecule-1 (sICAM-1), a marker of endothelial activation and inflammation, may improve early detection. This study evaluated the combined potential of VOC profiling and sICAM-1 to distinguish drug-sensitive TB (DS-TB), DR-TB, and healthy controls. Methods: A prospective cross-sectional study was conducted in four TB referral hospitals in East Java, Indonesia. Eighty-eight participants were enrolled: DS-TB (n=32), DR-TB (n=30), and healthy controls (n=26). Exhaled breath VOC profiles were analyzed using a breath analyzer, and serum sICAM-1 levels were measured by enzyme-linked immunosorbent assay. Group comparisons used ANOVA or Kruskal–Wallis tests with post hoc analyses, and correlations were assessed using Pearson or Spearman tests. Results: Serum sICAM-1 levels were significantly higher in both TB groups than in controls (p<0.01) but did not differ between DS-TB and DR-TB (p=0.961). Several VOC components differed significantly, particularly formaldehyde, acetone, and methane (CH₄). CH₄ levels were significantly higher in DR-TB than in DS-TB and controls (p<0.05). A weak negative correlation between formaldehyde and sICAM-1 was observed in DS-TB (r = −0.355; p = 0.046). Conclusion: Integrating VOC profiling with sICAM-1 assessment offers a promising non-invasive strategy for TB screening. While sICAM-1 reflects inflammatory activity, methane-based VOC signatures may improve identification of DR-TB.
- Open Access
- 8 April 2026
- Published by Universitas Pattimura in BAREKENG: Jurnal Ilmu Matematika dan Terapan
- Vol. 20 (3)
MULTIVARIATE ROBUST MONITORING OF PLASTIC WASTE QUALITY USING PCA-BAYESIAN MEWMA CONTROL CHART
Quality control is essential for ensuring manufacturing processes consistently meet predefined specifications and for minimizing the risks caused by process deviations. The MEWMA control chart is widely used for detecting small shifts in multivariate processes which does not require strict multivariate normality but the performance can be compromised when data contain outliers or high multicollinearity that commonly found in plastic waste processing. This study proposes a robust monitoring approach by integrating PCA to address multicollinearity, Bayesian estimation to improve parameter robustness. The four charts examined in this study are PCA-Bayesian MEWMA (SELF), PCA-Bayesian MEWMA (MSELF), PCA-Bayesian MEWMA (KLF), and PCA-MEWMA using Bootstrap control limit as comparison. These charts are evaluated across 324 simulated scenarios, varying in collinearity levels (0.2, 0.6, 0.95), sample sizes (10, 20, 30), outlier proportions (5%, 10%, 15%), and smoothing parameters (λ = 0.2, 0.5, 0.8). Performance is measured using Average Run Length (ARL), Standard Deviation of Run Length (SDRL), Median Run Length (MRL), and False Alarm Rate (FAR). Results indicate that the PCA-Bayesian MEWMA outperformed PCA-MEWMA using Bootstrap control limit. PCA-Bayesian MEWMA (SELF) excelled under clean data condition, whereas PCA-Bayesian (MSELF) provided stable detection under high correlation, moderate-to-high outlier contamination, and larger smoothing parameters, achieving an average ARL of 3.44, an SDRL of 0.58, an MRL of 3.46, and FAR of 0.03, making it well-suited for monitoring complex industrial plastic waste processes and demonstrating its effectiveness for robust quality monitoring in production.
- 8 April 2026
- Published by Universitas Pattimura in BAREKENG: Jurnal Ilmu Matematika dan Terapan
- Vol. 20 (3)
Open AccessMULTIVARIATE ROBUST MONITORING OF PLASTIC WASTE QUALITY USING PCA-BAYESIAN MEWMA CONTROL CHART
Quality control is essential for ensuring manufacturing processes consistently meet predefined specifications and for minimizing the risks caused by process deviations. The MEWMA control chart is widely used for detecting small shifts in multivariate processes which does not require strict multivariate normality but the performance can be compromised when data contain outliers or high multicollinearity that commonly found in plastic waste processing. This study proposes a robust monitoring approach by integrating PCA to address multicollinearity, Bayesian estimation to improve parameter robustness. The four charts examined in this study are PCA-Bayesian MEWMA (SELF), PCA-Bayesian MEWMA (MSELF), PCA-Bayesian MEWMA (KLF), and PCA-MEWMA using Bootstrap control limit as comparison. These charts are evaluated across 324 simulated scenarios, varying in collinearity levels (0.2, 0.6, 0.95), sample sizes (10, 20, 30), outlier proportions (5%, 10%, 15%), and smoothing parameters (λ = 0.2, 0.5, 0.8). Performance is measured using Average Run Length (ARL), Standard Deviation of Run Length (SDRL), Median Run Length (MRL), and False Alarm Rate (FAR). Results indicate that the PCA-Bayesian MEWMA outperformed PCA-MEWMA using Bootstrap control limit. PCA-Bayesian MEWMA (SELF) excelled under clean data condition, whereas PCA-Bayesian (MSELF) provided stable detection under high correlation, moderate-to-high outlier contamination, and larger smoothing parameters, achieving an average ARL of 3.44, an SDRL of 0.58, an MRL of 3.46, and FAR of 0.03, making it well-suited for monitoring complex industrial plastic waste processes and demonstrating its effectiveness for robust quality monitoring in production.
- Open Access
- 26 June 2025
- Published by Universitas Muhammadiyah Mataram in JTAM (Jurnal Teori dan Aplikasi Matematika)
- Vol. 9 (3) , 752-768
An Informative Prior of Bayesian Kriging Approach for Monthly Rainfall Interpolation in East Java
In spatial data analysis, interpolation is used to estimate values at unobserved locations, but often faces challenges in capturing complex spatial patterns and estimation uncertainty. One of the main obstacles is the small sample size, which makes the empirical variogram difficult to define well in conventional Kriging methods. The Bayesian Kriging approach overcomes this problem by integrating prior information, so it can still produce stable estimates despite limited data. This study is a quantitative, spatial-based research aimed at interpolating monthly rainfall in East Java Province using the Bayesian Kriging approach. The data consist of monthly rainfall measurements from 11 rain gauge stations distributed across East Java, obtained from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG) for the period of January to April 2024. The entire analysis was conducted using R software. A spherical semivariogram model was selected due to its superior fit to the spatial characteristics of the rainfall data in the study area with the smallest RMSE 37.17. This study demonstrates the effectiveness of Bayesian Kriging for rainfall interpolation in tropical regions with sparse data, providing more stable and accurate estimates compared to conventional methods. The scientific contribution of this research lies in showcasing how the integration of informative priors and Bayesian inference enhances interpolation accuracy in data-limited tropical environments. The resulting interpolated maps can inform land-use planning and flood risk mitigation by identifying areas of high rainfall for improved water infrastructure and lower-rainfall regions for targeted irrigation planning.
- 26 June 2025
- Published by Universitas Muhammadiyah Mataram in JTAM (Jurnal Teori dan Aplikasi Matematika)
- Vol. 9 (3) , 752-768
Open AccessAn Informative Prior of Bayesian Kriging Approach for Monthly Rainfall Interpolation in East Java
In spatial data analysis, interpolation is used to estimate values at unobserved locations, but often faces challenges in capturing complex spatial patterns and estimation uncertainty. One of the main obstacles is the small sample size, which makes the empirical variogram difficult to define well in conventional Kriging methods. The Bayesian Kriging approach overcomes this problem by integrating prior information, so it can still produce stable estimates despite limited data. This study is a quantitative, spatial-based research aimed at interpolating monthly rainfall in East Java Province using the Bayesian Kriging approach. The data consist of monthly rainfall measurements from 11 rain gauge stations distributed across East Java, obtained from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG) for the period of January to April 2024. The entire analysis was conducted using R software. A spherical semivariogram model was selected due to its superior fit to the spatial characteristics of the rainfall data in the study area with the smallest RMSE 37.17. This study demonstrates the effectiveness of Bayesian Kriging for rainfall interpolation in tropical regions with sparse data, providing more stable and accurate estimates compared to conventional methods. The scientific contribution of this research lies in showcasing how the integration of informative priors and Bayesian inference enhances interpolation accuracy in data-limited tropical environments. The resulting interpolated maps can inform land-use planning and flood risk mitigation by identifying areas of high rainfall for improved water infrastructure and lower-rainfall regions for targeted irrigation planning.
- Open Access
- 22 March 2025
- Published by Maulana Malik Ibrahim State Islamic University in CAUCHY: Jurnal Matematika Murni dan Aplikasi
- Vol. 10 (1) , 442-456
Geographically Weighted Random Forest Model for Addressing Spatial Heterogeneity of Monthly Rainfall with Small Sample Size
Random Forest (RF) is a flexible machine learning algorithm that does not rely on linearity. The use of the RF method for spatial analysis is referred to as Geographically Random Forest (GRF), which can capture the effects of spatial heterogeneity. Therefore, GRF is well-suited for modeling rainfall, which exhibits spatial heterogeneity characteristics. However, the limited sample size in rainfall data presents a challenge. The aim of this study is to apply and explore the performance of the GRF model in monthly rainfall modeling in East Java. The GRF model is applied with different numbers of trees in the forest to achieve an optimal model. The performance evaluation of the GRF model is assessed based on the smallest RMSE, AIC, and AICc values. The analysis results indicate that the model exhibits an overestimation of Out-of-Bag (OOB) Error across all variations in the number of trees, with the smallest RMSE obtained at 750 trees. Based on variable importance values, humidity is identified as the most important variable in determining monthly rainfall in East Java.
- 22 March 2025
- Published by Maulana Malik Ibrahim State Islamic University in CAUCHY: Jurnal Matematika Murni dan Aplikasi
- Vol. 10 (1) , 442-456
Open AccessGeographically Weighted Random Forest Model for Addressing Spatial Heterogeneity of Monthly Rainfall with Small Sample Size
Random Forest (RF) is a flexible machine learning algorithm that does not rely on linearity. The use of the RF method for spatial analysis is referred to as Geographically Random Forest (GRF), which can capture the effects of spatial heterogeneity. Therefore, GRF is well-suited for modeling rainfall, which exhibits spatial heterogeneity characteristics. However, the limited sample size in rainfall data presents a challenge. The aim of this study is to apply and explore the performance of the GRF model in monthly rainfall modeling in East Java. The GRF model is applied with different numbers of trees in the forest to achieve an optimal model. The performance evaluation of the GRF model is assessed based on the smallest RMSE, AIC, and AICc values. The analysis results indicate that the model exhibits an overestimation of Out-of-Bag (OOB) Error across all variations in the number of trees, with the smallest RMSE obtained at 750 trees. Based on variable importance values, humidity is identified as the most important variable in determining monthly rainfall in East Java.