1、西南交通大學(xué)碩士學(xué)位論文Copula回歸函數(shù)和方向相依姓名：苗迎樂申請學(xué)位級別：碩士專業(yè)：概率論與數(shù)理統(tǒng)計指導(dǎo)教師：何平20080401西南交通大學(xué)碩士研究生學(xué)位論文第H頁AbstractDepedenceisanimportantpartinprobabilityandstatiticsNowtheusedmeasurearePearsoncoralationKendal王’Sz，Spearman’SP，Buttheyareallnu

2、mbers，theycan’treflectthedependencerelationbeteenthe。randomvariablesexactly，Independencemodelingsuchasregression，oneattemptstoprovideafunctionalformthatwillsummarizetherelationshipbeteenresponseandexplanatoryThetypeoffun

3、ctionalrelationshipcoulddependonthemarginalbeohaviorsortheirjointbehaviorsInthesecondchapter，wewilllookatawayofseparatingthesebehaviors。WcwillreflecttherelationbyusingCopularegressionfunctionsInchapterthree，undertheassum

4、ptionoflinearrelationshipbeteentWOvariablesweprovidealternativesimplemethodofprovingtheexistingconnectingcorrelationcoefficientwiththoseofskewnessofresponseandexplanatoryvariables。Furtherwehavegivenarelationshipbetweenco

5、rrelationcoefficientandcoefficientofkurtosisofresponseandexplanatoryvariablesInchapterfour，wedefineandstudytheconceptofdirectionaldependenceinbivariateregressionsettingbyusingCopulaweconsidertwocaseofdirectional磊epende纛e

6、e：o鑫eoriginatingfrommarginalsandtheotheroriginatingfromjointbehaviorofvariableswealsogeneralizeandclarifytheresultgivenbychapterthree。wealsointroduceameasurethatcanbeemployedforbothtypesofdirectionaldependenceandprovidea