Adaptive Parameter Estimation of the Generalized Extreme Value Distribution Using Artificial Neural Network Approach

Article Computer Science, Interdisciplinary Applications

A user-friendly software for modelling extreme values: EXTRASTAR (EXTRemes Abacus for STAtistical Regionalization)

Davide Luciano De Luca et al.

Summary: The software EXTRASTAR is proposed as a useful tool for statistical analyses of Annual Maxima time series. It offers innovative features such as comparison of sample skewness, prediction intervals, and calibration of parameters, which help in obtaining accurate regional values for analysis. The software was successfully tested using the Italian rain-gauge network data.

ENVIRONMENTAL MODELLING & SOFTWARE (2023)

Add to Collection

Article Geochemistry & Geophysics

Analysis of maximum precipitation in Thailand using non-stationary extreme value models

Thanawan Prahadchai et al.

Summary: This study applies non-stationary models to analyze rainfall data from selected stations in Thailand, taking into account long-term climate trends. The results reveal increasing or decreasing trends in maximum precipitation at certain stations.

ATMOSPHERIC SCIENCE LETTERS (2023)

Add to Collection

Article Statistics & Probability

Penalized likelihood approach for the four-parameter kappa distribution

Nipada Papukdee et al.

Summary: The four-parameter kappa distribution (K4D) is widely applied in modeling in fields such as hydrology and climatic change. The estimation of the four parameters is usually done through the maximum likelihood approach and the method of L-moments. A maximum penalized likelihood estimation (MPLE) is proposed to improve on small sample properties by adjusting penalty functions that restrict the parameter space.

JOURNAL OF APPLIED STATISTICS (2022)

Add to Collection

Article Environmental Sciences

Future Projections and Uncertainty Assessment of Precipitation Extremes in the Korean Peninsula from the CMIP6 Ensemble with a Statistical Framework

Yonggwan Shin et al.

Summary: Scientists predict future changes in precipitation extremes on the Korean peninsula by utilizing multi-model ensemble methods and model weighting techniques to quantify uncertainty. The study suggests that by the end of the 21st century, there will be more prominent severe rainfall in the southern and central parts of the Korean peninsula.

ATMOSPHERE (2021)

Add to Collection

Article Environmental Sciences