Patchanok Srisuradetchai

1. The two-parameter Weibull distribution is a standard modeling tool in reliability engineering, survival analysis, hydrology, wind energy assessment, and applied risk modeling, owing to its flexi...

1. The Gumbel distribution is a classical model for block maxima and continues to play a central role in the statistical analysis of extreme events in hydrology, meteorology, reliability engineerin...

Likelihood-based inference for exponential-tail lifetime models can be unstable in finite samples, particularly under right-tail contamination. We develop a robust score-based inference procedure for the two-parameter Gompertz distribution by constructing estimating equations in which influence is bounded after standardisation by the local curvature of the likelihood, implemented via OPG-whitened score contributions and Huber weighting. Uncertainty is quantified using a Godambe (sandwich) covariance estimator, enabling Wald-type inference. Simulation results under clean sampling and controlled right-tail contamination show improved stability and more reliable coverage than maximum likelihood, with a moderate efficiency loss under correct specification.

This paper presents a nonparametric interval forecasting method that combines circular block bootstrap resampling with complexity-invariant K-nearest-neighbor time-series prediction. Prediction intervals are obtained directly from bootstrap-resampled training series, thereby preserving temporal dependence while accounting for forecast uncertainty. Under weak dependence and local stability assumptions, asymptotic validity of the resulting prediction intervals is established. The proposed method is evaluated using twelve time-series datasets drawn from economic, environmental, industrial, and energy applications. Empirical performance is compared with seasonal autoregressive integrated moving average models and long short-term memory neural networks using the mean absolute percentage error, empirical coverage probability, and interval score. The results show that the proposed approach yields prediction intervals of moderate width with competitive forecasting accuracy across most datasets, while empirical coverage remains close to the nominal level. Mild undercoverage is observed in short samples, attributable to limited data availability and fixed tuning parameters.

Naive Bayes (NB) classifiers are widely used for their simplicity, computational efficiency, and interpretability. However, their predictive performance can degrade significantly in real-world settings where the conditional independence assumption is often violated. More complex NB variants address this issue but typically introduce structural complexity or require explicit dependency modeling, limiting their scalability and transparency. This study proposes two lightweight ensemble-based extensions—randomized feature naive Bayes (RF-NB) and randomized feature bootstrapped naive Bayes (RFB-NB)—designed to enhance robustness and predictive stability without altering the underlying NB model. By integrating randomized feature selection and bootstrap resampling, these methods implicitly reduce feature dependence and noise-induced variance. Evaluation across twenty real-world datasets spanning medical, financial, and industrial domains demonstrates that RFB-NB consistently outperformed classical NB, RF-NB, and k-nearest neighbor in several cases. Although random forest achieved higher average accuracy overall, RFB-NB demonstrated comparable accuracy with notably lower variance and improved predictive stability specifically in datasets characterized by high noise levels, large dimensionality, or significant class imbalance. These findings underscore the practical and complementary advantages of RFB-NB in challenging classification scenarios.

The k-nearest neighbors (KNN) algorithm is widely recognized for its simplicity and flexibility in modeling complex, non-linear relationships; however, standard KNN regression does not inherently provide prediction intervals (PIs), presenting a persistent challenge for uncertainty quantification. This study introduces a bootstrap-based multi-K approach specifically designed to construct robust prediction intervals in KNN regression. By systematically aggregating predictions across multiple neighborhood sizes through ensemble techniques and bootstrap resampling, the method effectively quantifies prediction uncertainty, particularly in challenging high-dimensional scenarios. Evaluations conducted on 15 diverse datasets spanning education, healthcare, chemistry, economics, and social sciences reveal that the proposed approach consistently achieves competitive predictive accuracy compared to traditional regression methods. Although traditional regression produces wider intervals with higher coverage probabilities, the proposed bootstrap-based KNN method generates notably tighter intervals, enhancing interpretability and practical utility. Despite occasionally reduced coverage probabilities, especially in high-dimensional contexts, the proposed methodology effectively balances precision and predictive coverage. Practically, this multi-K bootstrap approach provides researchers and practitioners with an effective and interpretable method for robust uncertainty quantification in complex predictive modeling tasks.

Accurately forecasting sinusoidal time series is essential in various scientific and engineering applications. However, traditional models such as the seasonal autoregressive integrated moving average (SARIMA) rely on assumptions of linearity and stationarity, which may not adequately capture the complex periodic behaviors of sinusoidal data, including varying amplitudes, phase shifts, and nonlinear trends. This study investigates Monte Carlo dropout neural networks (MCDO NNs) as an alternative approach for both forecasting and uncertainty quantification. The performance of MCDO NNs is evaluated across six sinusoidal time series models, each exhibiting different dynamic characteristics. Results indicate that MCDO NNs consistently outperform SARIMA in terms of root mean square error, mean absolute percentage error, and the coefficient of determination, while also producing more reliable prediction intervals. To assess real-world applicability, the airline passenger dataset is used, demonstrating MCDO’s ability to effectively capture periodic structures. These findings suggest that MCDO NNs provide a robust alternative to SARIMA for sinusoidal time series forecasting, offering both improved accuracy and well-calibrated uncertainty estimates.

Deep neural networks (DNNs) are prominent in predictive analytics for accurately forecasting target variables. However, inherent uncertainties necessitate constructing prediction intervals for reliability. The existing literature often lacks practical methodologies for creating predictive intervals, especially for time series with trends and seasonal patterns. This paper explicitly details a practical approach integrating dual-output Monte Carlo Dropout (MCDO) with DNNs to approximate predictive means and variances within a Bayesian framework, enabling forecast interval construction. The dual-output architecture employs a custom loss function, combining mean squared error with Softplus-derived predictive variance, ensuring non-negative variance values. Hyperparameter optimization is performed through a grid search exploring activation functions, dropout rates, epochs, and batch sizes. Empirical distributions of predictive means and variances from the MCDO demonstrate the results of the dual-output MCDO DNNs. The proposed method achieves a significant improvement in forecast accuracy, with an RMSE reduction of about 10% compared to the seasonal autoregressive integrated moving average model. Additionally, the method provides more reliable forecast intervals, as evidenced by a higher coverage proportion and narrower interval widths. A case study on Thailand’s durian export data showcases the method’s utility and applicability to other datasets with trends and/or seasonal components.