Evaluation of Sampling Properties of General Extreme Value (GEV) Distribution-L-Moments Vs Conventional Moments

by A. Sankarasubramanian,
K. Srinivasan,

Document Type: Proceeding Paper

Part of: North American Water and Environment Congress & Destructive Water


The sample estimates of conventional moments are shown to be biased and this bias increases for small samples, and especially so when the skewness is high (Wallis et al., 1974). The recently introduced L-moments (Hosking, 1990) (analogous to conventional moments) are quoted to be unbiased, comparatively, even though the sampling properties of L-moments have not yet been investigated in detail. It is of interest in this paper, to investigate and compare the sampling properties of L moments and Conventional moments for the General Extreme Value (GEV) distribution, a popular flood frequency distribution, often adopted in Regional Flood Frequency Analysis.

Subject Headings: Frequency distribution | Flood frequency | Statistics | Frequency analysis | Skewness | Regional analysis | Droughts

Services: Buy this book/Buy this article


Return to search