과산포 현상을 고려한 가구내 비가정기반통행발생 모형구축

• 영문명: Modeling Non-Home Based Trip Generation of Household Considering Overdispersion
• 발행기관: 한국지역개발학회
• 저자명: 장태연(Jang, Tae Youn) 김대영(Kim Dae Young) 김정호(Kim Jung Ho) 권진영(Kwon Jin Young)
• 간행물 정보: 『한국지역개발학회지』제14권 제3호, 123~134쪽, 전체 12쪽
• 주제분류: 사회과학 > 지역개발
• 파일형태: PDF
• 발행일자: 2002.12.30

영문 초록

Trip generation in fotrr sP trav， fα-ecasting rmdel uses lear rression rrx:x However, 1eaI" I'Eession m:x has several invalid characlstics. First, trip generation frequencies generally show discrete distribution but the rrx:xEl is esnated sed on the assurrφdon that dtgent variable has tl1e norma1 distribution. secαld， tl1e mαel may be able to provide fiiative vues of (k~dent variable. Third, even if the forecasted v，ue is within valid ranges, only trip frequencies as esψnated rest n offered but oot di-ete probabty distribution To overcorre tl1ese shortcomings, ∞unt data rrxxEls such as poisson rression or native binomial rression can be Poissα1 rression mαiel with a assunon of uidispersion s fra}uently been :rl to 1alyze ∞oot data However, if there is overdis"SÌon in data pois1 1ression mα ds to UIerestimates errors, res19 in α"Oblem in r머iab피ty. In the study applying lÌs ∞ncept into analysis on oon-horre basc;p generation in houslo1d， overdispersion testing, it tries to find the possibility to use n(ative binomial rression rrx:xEl ∞nsidtng overdispersion and irr"Oving the iability. Also, through re1αxi test and πleil Irlity Coefficient test, the m:x:lc: IS Cclred for i reliability and accura.

목차

1. 서론

2. 가산자료 모형

3. 모형의 추정

4. 결론

참고문헌

키워드

가산자료(count data) 포와송회귀모형 (poisson regression mode)음이항회귀 모형(negative binomia1 regression model) 과산포 (overdispersion)