• Title/Summary/Keyword: Quanitle Regression

Search Result 1, Processing Time 0.017 seconds

A Study on Trend Analysis in Sea Level Data Through MK Test and Quantile Regression Analysis (MK 검정 및 분위회귀분석을 통한 해수면 자료의 경향성 평가에 관한 연구)

  • Uranchimeg, Sumiya;Kim, Yong-Tak;Kwon, Hyun-Han;Hwang, Kyu-Nam
    • Journal of Korean Society of Coastal and Ocean Engineers
    • /
    • v.27 no.2
    • /
    • pp.94-104
    • /
    • 2015
  • Population and urban development along the coast is growing in South Korea, and particularly sea level rise is likely to increase the vulnerability of coastal areas. This study aims to investigate the sea level rise through Mann-Kendall(MK) test, ordinary linear regression(OR) and quantile regression analysis(QRA) with sea level data at the 20 tide stations along the coast of Korean Peninsula. First, statistically significant long-term trends were analysed using a non-parametric MK test and the test indicated statistically significant trends for 18 and 10 stations at the 5% significance level in the annual mean value of sea level and the annual maximum value of sea level, respectively. The QRA method showed better performance in terms of characterizing the degree of trend. QRA showed that an average annual rise in mean sea level is about 1-6 mm/year, and an average rise in maximum sea level is about 1-20 mm. It was found that upward convergent and upward divergent were a representative change given the nine-category distributional changes. We expect that in future work we will address nonstationarities with respect to sea level that were identified above, and develop a nonstationary frequency analysis with climate change scenarios.