完整後設資料紀錄
DC 欄位語言
dc.contributor.author鄭佳瑜en_US
dc.contributor.authorChia-Yu Chengen_US
dc.contributor.author洪志真en_US
dc.contributor.author洪慧念en_US
dc.contributor.authorDr.Jyh-Jen Horng Shiauen_US
dc.contributor.authorDr.Hui-Nien Hungen_US
dc.date.accessioned2014-12-12T02:30:08Z-
dc.date.available2014-12-12T02:30:08Z-
dc.date.issued2002en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#NT910337007en_US
dc.identifier.urihttp://hdl.handle.net/11536/70037-
dc.description.abstract在很多資料分析中,都有所謂的因果關係 (cause-effect relationship),也有另一種說法是親-子關係(parent-children relationship) ,例如身高與體重,身高增加一般可以讓體重自然地加重,但是加重體重卻無法達到長高的目的,這例子中身高就是因,體重則為果。若是基因間的關係,很多基因間都有活化(inspire/active)或抑制(depress/repress)的機制,A基因(parent)能活化B基因(child),並不代表B基因可以抑制或活化A基因。在不知道哪個變數才是親輩的情況下,如何衡量這樣的關係而且找出正確的親輩,一直就是現在科學家想探討的部分。而本文主要的目標是希望能從這樣有因果關係的資料主要是找出其方向性。 針對屬量的資料,之前的一些文獻使用貝氏網路 (Bayesian Network)方法去衡量變數之間的關係。本文則是提出一個使用無母數迴歸的方法來衡量資料的因果關係, 並討論其用於因果關係上的特性與限制zh_TW
dc.description.abstractConsider the problem of determining the cause-effect relationship between two random variables X and Y from n pairs of independent and identically distributed data {(xi, yi), i=1,…,n}. Functional causal model with a causal graph is adopted to describe the causal relationship. The functional relationship is estimated by the smoothing spline estimation method. Between the two causal graph candidates, “X→Y” and “Y→X”, we propose a selection criterion based on a Bayesian approach. A simulation study is conducted to evaluate the performance of the method. Results are promising. Some factors affecting the performance of the method are discussed.en_US
dc.language.isozh_TWen_US
dc.subject因果關係zh_TW
dc.subject無母數迴歸zh_TW
dc.subjectCause-Effecten_US
dc.subjectNonparametric Regressionen_US
dc.subjectparent-children relationshipen_US
dc.title用無母數迴歸法分析變數間之因果關係zh_TW
dc.titleDetermining the Cause-Effect Relationship between Two Variables by Nonparametric Regressionen_US
dc.typeThesisen_US
dc.contributor.department統計學研究所zh_TW
顯示於類別:畢業論文