METODE BAYESIAN PADA MODEL REGRESI LINEAR

JATU P, TRIWIK (2007) METODE BAYESIAN PADA MODEL REGRESI LINEAR. Other thesis, Universitas Sebelas Maret.

[img] PDF - Published Version
Download (340Kb)

    Abstract

    ABSTRACT Triwik Jam Parmanmgh. 2037. BAYESIAN ?ðETH CD ON LINEAR RE GRESSION MODEL. Faculty of Mathematics and Natuml Sciences. Sebelas Man’t Univemity. Regreaicii aiial y145 IS a statistical technique fkr invHtigating the relation diip l)etweeu variables. The least squaw itiethod is cL,11111111ly iii’d to estimate wgn’sion paniua’ten. Srnwtinws. when ilniwing an inft’it’nce 1mm population we get an additional in formation abon t parnxilcters avail ab le 1mm n previous study. If the pievious inf(nnIation will be incorporated in data analysis. then the esti ¡nation by t he least *Iuan iiut hod can not be ud. Therefore, we need Bayeaii method to dve sida problem. Tue aims of this research are to estimate regresiion pammvten. to test igniÍicance of n’gn’sion pammetem by Bayesian hat hod and to compare th e est ¡mat ion of regreiiohi model based on noam lumia— tive prior (Iistnbuturn and conjugate pnor distribution in a ca’e. The itiethod used in this research is literary studv. The steps are to explain Bavei4an method, to estimate reel4(.l pamuiwt ers by the mean of posterior dis— tribut ion, to detenitine pc.4terior pmbability For test tug ¡4gni Ilcance of re’easion pamnwten4 and to analyze the ermr to compare the estimation of wgwsion ¡nu— (leI bared on noninfurnintive prior (liSt nbutic,i and conjugate pnchm (liatributiun in a case. This research indicates that Ban’s estilitation turns out to be ectly the as the least iiiare estimate if pnor information is not known, whereas if prior information is known then Bayes estimation is taken to be the expectation of postenor distribution. The hypothe,is test in Bayesian method is based on postenor pmbability calculation Prr)J(JIiìIy). The Bayesian test rejects the null hypotesis if Prd(Hi,Iy < PrdiHo) Key word8: prior disrthutioya. po.utertir djqfributian. Boijc.a rstàndor. iv

    Item Type: Thesis (Other)
    Subjects: Q Science > QA Mathematics
    Divisions: Fakultas Matematika dan Ilmu Pengetahuan Alam > Matematika
    Depositing User: Saputro Bagus
    Date Deposited: 13 Jul 2013 05:25
    Last Modified: 13 Jul 2013 05:25
    URI: https://eprints.uns.ac.id/id/eprint/2983

    Actions (login required)

    View Item