<?xml version="1.0" encoding="UTF-8" ?>
<modsCollection xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://www.loc.gov/mods/v3" xmlns:slims="http://slims.web.id" xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-3.xsd">
<mods version="3.3" id="32002">
 <titleInfo>
  <title>Deteksi Outlier pada fungsi transfer menggunkan metode Cook's Squared Distance, Dffits dan Dfbetas</title>
 </titleInfo>
 <name type="Personal Name" authority="">
  <namePart>Koriatun</namePart>
  <role>
   <roleTerm type="text">Primary Author</roleTerm>
  </role>
 </name>
 <typeOfResource manuscript="no" collection="yes">mixed material</typeOfResource>
 <genre authority="marcgt">bibliography</genre>
 <originInfo>
  <place>
   <placeTerm type="text">Bandung</placeTerm>
   <publisher>Magister Statistika Terapan</publisher>
   <dateIssued>2010</dateIssued>
  </place>
 </originInfo>
 <language>
  <languageTerm type="code">id</languageTerm>
  <languageTerm type="text">Indonesia</languageTerm>
 </language>
 <physicalDescription>
  <form authority="gmd"></form>
  <extent>xviii,; 82 hlm,;29 cm</extent>
 </physicalDescription>
 <note>Fungsi transfer merupakan kasus khusus regresi linier yang diterapkan pada &#13;
data deret waktu. Hal ini disebabkan deret gangguan (noise) pada fungsi transfer &#13;
diasumsikan sama seperti gangguan (error) pada model regresi linier, yaitu &#13;
bersifat independent identically distributed mengikuti distribusi normal dengan &#13;
rata-rata nol dan varians konstan. Perbedaannya bahwa variabel input, variabel &#13;
output dan deret gangguan pada fungsi transfer mempunyai autokorelasi, &#13;
sedangkan variabel prediktor, variabel respon dan error pada model regresi linier &#13;
tidak berautokorelasi. Oleh karena itu pada model fungsi transfer hams dilakukan &#13;
pemutihan deret (pre-whitening) untuk menghilangkan pola yang ada. &#13;
&#13;
Berdasarkan asumsi tersebut, maka pada penelitian ini diterapkan metode &#13;
deteksi outIier menggunakan Cook' Squared Distance, DFFITS dan DFBETAS &#13;
pada fungsi transfer. Penelitian ini bertujuan untuk mengukur kelayakan &#13;
penerapan ketiga metode tersebut pada fungsi transfer. Data yang digunakan &#13;
adalah data skunder yang bersumber dari Bank Indonesia sejak Januari 2001 &#13;
sampai Maret 2010. Metode Analisis yang digunakan adalah identifikasi model &#13;
ARIMA (p,d,q) dan pembentukan model fungsi transfer. Selanjutnya dari model &#13;
fungsi transfer yang diperoleh dilakukan deteksi outIier menggunakan Cook's &#13;
Squared Distance, DFFITS dan DFBETAS. Hasil penelitian menunjukan bahwa &#13;
metode DFFITS dan DFBETAS lebih layak diterapkan pada model fungsi transfer &#13;
bila dibandingkan metode Cook's Squared Distance. &#13;
&#13;
Abstract &#13;
&#13;
Transfer function is a special case of linear regression which it is applied in &#13;
time series analysis, because the noise in transfer function has a same &#13;
assumption with error on linear regression models,' both of them have an &#13;
independent identically distributed, i.e. normal distribution with zerro mean and &#13;
constant variance. The difference is the input variable, output variable and the &#13;
&#13;
IV &#13;
&#13;
noise on transfer function model have autocorrelation, while the predictor &#13;
variable, response variable and error in linear regression didn't have an &#13;
autocorrelation. Therefore, the pre-whitening of transfer function must be made &#13;
to remove the existing pattern. &#13;
&#13;
Based on those assumptions, the outlier detection method using the Cook's &#13;
Squared Distance, DFFITS and DFBETAS to the transfer function are applied to &#13;
this study. The purpose of this study was to measure the feasibility of applying &#13;
those methods on the transfer function. The data used in this study are secondary &#13;
data obtained from the Bank Indonesia since January 2001 to March 2010. The &#13;
analysis methods used the identification of ARIMA (p,d,q) and the transfer &#13;
function models. Furthermore, the transfer function model can be detect on &#13;
outlier using the Cook's Squared Distance, DFFITS and DFBETAS. So it can be &#13;
concluded that the DFFITS and DFBETAS are more feasible to be applied on &#13;
transfer function than the Cook's Squared Distance. &#13;
&#13;
</note>
 <note type="statement of responsibility">Koriatun</note>
 <subject authority="">
  <topic>Fungsi Transfer, ARIMA, Cook's Squared Distance,  </topic>
 </subject>
 <classification>519.5 Kor d</classification>
 <identifier type="isbn"></identifier>
 <location>
  <physicalLocation>Perpustakaan Universitas Padjadjaran Kementerian Riset Teknologi dan Pendidikan Tinggi</physicalLocation>
  <shelfLocator>519.5 Kor d/R.14.45</shelfLocator>
  <holdingSimple>
   <copyInformation>
    <numerationAndChronology type="1">01001100700088</numerationAndChronology>
    <sublocation>Perpustakaan Pusat (REF.14.45)</sublocation>
    <shelfLocator>519.5 Kor d/R.14.45</shelfLocator>
   </copyInformation>
  </holdingSimple>
 </location>
 <slims:image>20%252Fscan0001.jpg.jpg</slims:image>
 <recordInfo>
  <recordIdentifier>32002</recordIdentifier>
  <recordCreationDate encoding="w3cdtf">2017-04-02 00:52:59</recordCreationDate>
  <recordChangeDate encoding="w3cdtf">2018-05-23 10:13:15</recordChangeDate>
  <recordOrigin>machine generated</recordOrigin>
 </recordInfo>
</mods>
</modsCollection>