English  |  正體中文  |  简体中文  |  Items with full text/Total items : 52068/87197 (60%)
Visitors : 8905809      Online Users : 309
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/77484

    Title: 不同機率分布下特異值檢定之門檻值研究
    Other Titles: A study on the threshold of outlier detection for different probability distributions
    Authors: 邱薇如;Chiu, Wei-Ju
    Contributors: 淡江大學水資源及環境工程學系碩士班
    虞國興;Yu, Gwo-Hsing;鄭思蘋;Cheng, Szu-ping
    Keywords: 特異值;漢佩爾辨識法;機率分布;Outlier;Hampel identifier;Probability ditribution
    Date: 2012
    Issue Date: 2012-06-21 06:49:04 (UTC+8)
    Abstract:   近年來全球氣候變化甚大,極端洪水現象日漸加劇,然而,這些極端降雨事件是否為單一個案,亦或是氣候變異之徵兆,需進一步分析了解。進行水文分析時,若將極端降雨事件皆視為異常觀測值,進而將之刪除,則會造成資料分析結果不正確,以致無法準確判斷其降雨分布之型態,亦會造成災情模擬時的偏估狀況。
      在過去研究中,漢佩爾辨識法(Hampel identifier method)只用於對稱頻率分布資料,尚無探討非對稱機率分布。因此,本研究推導出非對稱機率分布之尺度因子K值,並依照不同分布提出有效之理論K值迴歸方程式,進一步計算偵測特異值之門檻值,並利用台灣南部地區之實測資料,以探討特異值偵測狀況,實測資料分別以短延時1小時及長延時24小時之時雨量進行分析。此外,在分析過程中發現各種降雨特性在近二十年來有所之改變,故本研究進一步將1990年作為之實測雨量資料之分界點進行探討。
      由研究結果顯示,漢佩爾辨識法可用於非對稱機率分布,只需得知其轉換尺度因子K值,即可應用於實際資料中;實際應用建議以α= 0.01為此法之門檻值標準;此外,以1990年分段後所偵測出之特異值結果不同,近年颱風之異常降雨被偵測為特異值結果較少,故台灣降雨情形於1990年前後確實有氣候變異之情形發生。
    Analysis is required to determine whether what appeared to be an increase in flood and extreme weather events signifies a global climate shift. When extreme weather events are excluded, the hydrological analysis would not yield correct conclusions about precipitation patterns or accurately predict the possible magnitudes of future climate disasters.
    The Hempel identifier method has mostly been applied to time series with symmetrical distributions. In this study, the Hempel identifier method was applied to time series of several skewed distributions, and a regression function was proposed to model the K-value for each type of skewed distribution. For the study area of southern Taiwan, the Hempel identifier method was applied to detect outlier values in 1-hr and 24-hr accumulated rainfall data. Also, a shift from historical patterns was detected in the recent rainfall data. Therefore, the rainfall data was divided into two sets, one prior to 1990 and one starting in 1990, to be analyzed separately for comparison.
    The results showed Hampel identifier method can be applied to time series of skew distributions. A best fit regression function relating K-value to the skewness of the data was determined using adjusted R-square values. Prior to 1990, extreme rainfalls brought by typhoons were often identified as outliers. In comparison, since 1990, a smaller proportion of extreme rainfalls brought by typhoons were identified as outliers, signifying a shift in climate pattern.
    Appears in Collections:[水資源及環境工程學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat

    All items in 機構典藏 are protected by copyright, with all rights reserved.

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback