BROWSE
Detail
Journal Article
Generalized method of L-moment estimation for hydrological data
Cited - time in 
Share 
Share
- Authors
- Yonggwan Shin, Yire Shin, Jihong Park, Jeong-Soo Park
- Citation
- Hydrological Sciences Journal, Early Access
- ISSN
- 0262-6667
- Publisher
- Taylor & Francis
- Language
- English
- Type
- Journal Article
- Abstract
- Reliable estimation of upper quantiles for long return periods is important in flood frequency analysis. The generalized extreme value (GEV) distribution is one of the most widely used models for this purpose. GEV parameters are commonly estimated by maximum likelihood estimation (MLE) or L-moment estimation (LME). Several studies have extended MLE by incorporating penalty functions or prior information. However, a generalized LME for the same purpose has not yet been developed. We therefore propose a generalized L-moment estimation (GLME) method. The proposed method is applied to stationary and nonstationary GEV models. Simulation results show that GLME reduces the bias of return level estimates compared with LME. Applications to two hydrological datasets demonstrate its practical usefulness.
- Keyword
- Bias correction, data-adaptive penalty function, generalized L-moment distance, rainfall extremes, return level
- KSP Keywords
- Adaptive penalty function, Bias Correction, Data-adaptive, Flood frequency analysis, Generalized extreme value (GEV) distribution, Hydrological data, Moment estimation, maximum likelihood estimation(MLE), prior information, reliable estimation, simulation results