Generalised extreme value model with cyclic covariate structure for analysis of non-stationary hydrometeorological extremes

Journal of earth system science

Jagtap R S^1¹  Gedam V K^2² 

[1] Central Water and Power Research Station, Government of India, Khadakwasla, Pune 411 024, India.^1;Department of Statistics & Centre for Advanced Studies in Statistics, Savitribai Phule Pune University, Pune 411 007, India.^2

关键词: Covariate;    cyclicity;    extreme values;    trend;    quantile;   

DOI  :  

学科分类：天文学(综合)

来源: Indian Academy of Sciences

【 摘 要 】

Studies carried out recently on hydrometeorological extremes report the evidence of non-stationarity induced by potential long-term climatic fluctuations and anthropogenic factors. A critical examination of the stationarity assumption has been carried out and a non-stationary generalised extreme value model with cyclic covariate structure for modelling magnitude and variation of data series with some degrees of correlation for real-world applications is proposed. Interestingly, the sinusoidal function with periodicity around 30 yr has been derived as a suitable covariate structure to deal with the ambiguous nature of temporal trends and this could possibly be linked to ‘Sun cycles’. It has adequately explained the cyclic patterns recognised in the annual rainfall which are helpful for realistic estimation of quantiles. Various diagnostic plots and statistics support the usefulness of the proposed covariate structure to tackle potential non-stationarities in the data characterising extreme events in various fields such as hydrology, environment, finance, etc.

【 授权许可】

CC BY