cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota bogor,
Jawa barat
INDONESIA
FORUM STATISTIKA DAN KOMPUTASI
Published by Institut Pertanian Bogor
ISSN : 08538115     EISSN : -     DOI : -
Core Subject : Education,
Mathematics
Forum Statistika dan Komputasi (ISSN:0853-8115) was published scientific papers in the area of statistical science and the applications. It is issued twice in a year. The papers should be research papers with, but not limited to, following topics: experimental design and analysis, survey methods and analysis, operation research, data mining, statistical modeling, computational statistics, time series and econometrics, and statistics education.
Arjuna Subject : -
Articles 119 Documents
 2   3   4   5   6 
PENDUGAAN REGESI SEMIPARAMETRIK DENGAN PENDEKATAN MODEL CAMPURAN LINEAR Anik Djuraidah
FORUM STATISTIKA DAN KOMPUTASI Vol. 14 No. 2 (2009)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (479.934 KB)

Abstract

Hubungan fungsional antara respon dengan peubah penjelas pada  regresi linear berganda berbentuk parametrik dengan metode pendugaan parameternya adalah metode kuadrat terkecil. Pada regresi semiparametrik, hubungan  fungsional antara respon dengan peubah penjelas dapat berbentuk parametrik atau nonparametrik. Metode yang banyak digunakan untuk pendugaan regresi semiparametrik adalah algoritma backfitting yang dikemukakan oleh Hastie & Tibshirani (1990). Pada penelitian ini pendugaan regresi parametrik didekati dengan model campuran linear. Keuntungan utama pendekatan  dengan model campuran linear adalah menggunakan metode ML atau REML sehingga memberi kemudahan dalam seleksi model dan penarikan kesimpulan
PEMODELAN KALIBRASI PEUBAH GANDA DENGAN PENDEKATAN REGRESI SINYAL P-SPLINE . Tonah; Ahmad Ansori Mattjik; Khairil Anwar Notodiputro
FORUM STATISTIKA DAN KOMPUTASI Vol. 14 No. 2 (2009)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (936.118 KB)

Abstract

Model kalibrasi peubah ganda merupakan suatu fungsi hubungan antara satuan pengukuran yang dapat diperoleh melalui proses yang relatif mudah atau murah dengan satuan pengukuran yang memerlukan waktu lama dan biaya mahal. Secara umum data kalibrasi memiliki multikolinearitas yang tinggi antar peubah penjelas dan dimensinya jauh lebih besar daripada banyaknya contoh yang dimiliki. Oleh karena itu, sebagian besar pendekatan model kalibrasi memerlukan pereduksian data terlebih dulu. Solusi alternatif bagi pemodelan kalibrasi adalah regresi sinyal P-spline (RSP). RSP merupakan salah satu pendekatan nonparametrik yang mensyaratkan bahwa koefisien regresi berada dalam ruang fungsi mulus. Hal ini dilakukan dengan cara merepresentasikan koefisien regresi sebagai kombinasi linear dari basis B-spline. Penambahan penalti dilakukan untuk mengatasi multikolinearitas pada model serta meningkatkan kemulusan koefisien regresi. Indeks dari bilangan gelombang yang terukur oleh FTIR digunakan sebagai domain B-spline. Spektra gingerol diidentifikasi memiliki pengaruh pencaran multiplikatif, sehingga perlu dilakukan koreksi pencaran. Model RSP dengan koreksi pencaran multiplikatif pada senyawa aktif gingerol memberikan hasil prediksi yang lebih baik. Hal ini ditunjukkan oleh nilai RMSEP dan R2y vs ŷ masing-masing sebesar 0.06867 dan 95.73 %. Nilai-nilai tersebut jauh lebih kecil dari hasil yang diberikan oleh model regresi komponen utama dengan pra-pemrosesan koreksi pencaran maupun transformasi wavelet.
METODE PENDUGAAN MATRIKS RAGAM-PERAGAM DALAM ANALISIS REGRESI KOMPONEN UTAMA (RKU) Itasia Dina Sulvianti; Dian Kusumaningrum; Yani Suryani
FORUM STATISTIKA DAN KOMPUTASI Vol. 14 No. 2 (2009)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (583.386 KB)

Abstract

Regresi komponen utama (RKU) merupakan salah satu analisis regresi yang menggunakan komponen utama untuk mengatasi adanya multikolinearitas pada regresi berganda. Metode kemungkinan maksimum (MLE) biasanya digunakan untuk menduga matrik ragam-peragam pada analisis regresi komponen utama. Namun, metode pendugaan ini sangat sensitif terhadap adanya data pencilan multivariat. Oleh karena itu, salah satu cara untuk mengatasi masalah ini adalah dengan menggunakan metode minimum covariance determinant (MCD) dalam menduga matriks ragam-peragamnya. Penelitian ini menggunakan metode MLE dan MCD untuk menduga matriks ragam-peragam pada analisis regresi komponen utama. Sedangkan parameter regresinya diduga oleh metode kuadrat terkecil (MKT). Sementara itu, untuk pemilihan jumlah komponen utama digunakan  kriteria 80% proporsi keragaman dari data contoh. Hasil penelitian ini menunjukkan bahwa dampak adanya pencilan multivariat pada analisis regresi komponen utama yang matriks ragam-peragamnya diduga oleh metode MCD akan menghasilkan nilai rata-rata akar ciri pertama yang tetap stabil pada komponen utama pertama (KU1), walaupun rasio pencilan multivariat dengan banyaknya data terus bertambah. Saat rasio pencilan multivariat dengan banyaknya data sebesar 5%, metode pendugaan parameter regresi komponen utama dengan MKT-MLE dan MKT-MCD menunjukkan hasil yang sama baik karena kedua metode ini cenderung menghasilkan nilai bias dan mean squared error (MSE) yang relatif sama kecil. Namun, pada saat rasio pencilan multivariat dengan banyaknya data lebih besar dari 5% (10%,15%,20%), metode MKT-MCD menunjukkan hasil yang lebih baik dibandingkan metode MKT-MLE dalam menduga parameter regresi komponen utama. Hal ini terjadi karena metode MKT-MCD cenderung menghasilkan nilai bias dan MSE yang lebih kecil dibandingkan MKT-MLE.
PENGGUNAAN ALGORITMA SIMULATED ANNEALING UNTUK MENYELESAIKAN TEKA-TEKI BINARY DAN SUDOKU ( Solving Binary and Sudoku Puzzles with a Simulated Annealing Algorithm ) Bagus Sartono
FORUM STATISTIKA DAN KOMPUTASI Vol. 16 No. 2 (2011)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (257.631 KB)

Abstract

Binary and Sudoku puzzles could be seen as optimization problems by using a score  of  rules  violation  as  the  objective  function  which  is  minimized. The simulated annealing algorithm is a good alternative to solve the puzzles.  This paper  describes the  approach  which  implements  the  algorithm  and  presents the  SAS/IML  program  of  it.    Empirical  trials  show  that  the  approach  works well to find the solution of the puzzles in a satisfying run time.  Keywords : meta-heuristic, simulated annealing
MODEL VEKTOR AUTOREGRESSIVE UNTUK PERAMALAN CURAH HUJAN DI INDRAMAYU (Vector Autoregressive Model for Forecast Rainfall In Indramayu ) Dewi Retno Sari Saputro; Aji Hamim Wigena; Anik Djuraidah
FORUM STATISTIKA DAN KOMPUTASI Vol. 16 No. 2 (2011)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (426.491 KB)

Abstract

There  are  three  regions  of  rainfall  that  has  been  formed,  each  rainfall  regions has a variety of homogeneous and there is a correlation between rainfall stations. In  each  region  can  be determined  rainfall  prediction  model simultaneously.  The model  is  a  model  of  Vector Autoregressive  (VAR)  which  is  an extension  of  the autoregressive  model  (AR).  Based  on  this  research,  we  can  determine  the  VAR model by lag 1 or VAR (1) for each region. Region 1 (Anjatan and Sumurwatu), region  2  (Salamdarma  and  Gantar)  and  region  3  (Kedokan  Bunder  and Sudimampir), each of which has a Root Mean Square Error Prediction (RMSEP) of  3.93;  5:03;  4:48;  5.3;  2:18  and  3:53.  Correlation  value  of  observations  with predictions of rainfall respectively, 0.71; 0.62; 0:57; 0:59; 0.89, and 0.91.  Keywords: AR, VAR, RMSEP, correlation
REGRESI KUADRAT TERKECIL PARSIAL MULTI RESPON UNTUK STATISTICAL DOWNSCALING (Multi Response Partial Least Square for Statistical Downscaling) Aji Hamim Wigena
FORUM STATISTIKA DAN KOMPUTASI Vol. 16 No. 2 (2011)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (203.349 KB)

Abstract

In  climatology  partial  least  square  regression  (PLSR)  can  be  used  as  an alternative  technique  in  statistical  downscaling  based  on  global  circulation model  (GCM)  output.  PLSR  is  the  technique  to  forecast  not  only  one  response but  also  multi  responses  to  accommodate  the  correlation  among  responses. PLSR is compared to PCR (Principal Component Regression). The results show that PLSR is better than PCR and can be used to forecast rainfall simultaneously in more than one rainfall stations relatively as well as in one station.  Keywords: statistical downscaling, PLSR, PCR, multi responses
PENDETEKSIAN PERILAKU HERDING PADA PASAR SAHAM INDONESIA DAN ASIA PASIFIK (Detection of Herding Behavior on Indonesia and Asia Pacific Stock Market) . Gunawan; Hari Wijayanto; Noer Azam Achsani; La Ode Abdul Rahman
FORUM STATISTIKA DAN KOMPUTASI Vol. 16 No. 2 (2011)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (461.889 KB)

Abstract

Herding  Behavior  is  an  irrational  investor  behavior, because  investors  do  not make investment decisions based on economic fundamentals of  risky assets, but based  on  others  investor  in  the  same  condition,  or  following  market  consensus. Herding  behavior  indications  can  be  seen  from  relation  between  dispersion  of stock  return  (Cross  Sectional  Absolute  Deviation,  CSAD)  and  market  portfolio return. If herding behavior exist, CSAD increases lower than increase of market portfolio  return  moreover,  CSAD  will  decrease  even  though  market  portfolio return increases. Herding behavior in stock market can trigger mislead in stock pricing  because  is  bias  among  investors  in  analyzing  risk  and  return.  To understand  relationship  between  CSAD  and  market  portfolio  return  in  some conditions, Quantil regression is used. Result gained from this research is that in Indonesian  and  global  Asia  Pacific  stock  market,  herding  behavior  occurs  in  a market stress condition, whereas in normal condition or in condition of very high stock return, investor behavior tends to be more rational.  Keywords : herd behavior, Quanrtile regression, CSAD
KAJIAN SIMULASI KETAKNORMALAN PENGARUH ACAK DAN BANYAKNYA DERET DATA LONGITUDINAL DALAM PEMODELAN BERSAMA (JOINT MODELING) (Simulation Study of Random Effects Nonnormality and Number of Longitudinal Data Series in Joint Modeling) Indahwati .; Aunuddin .; Khairil Anwar Notodiputro; I Gusti Putu Purnaba
FORUM STATISTIKA DAN KOMPUTASI Vol. 16 No. 2 (2011)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (335.827 KB)

Abstract

Joint modeling is intended to model longitudinal response process that affect the other primary response based on  assumption that both  processes induced by the same random effects. One of the assumptions that must be met in joint modeling is  normality  of  random  effects  and  intra-subject  error.  The  simulation  results show that the robustness of parameter estimates of joint model to the assumption of  random  effects  normality  can  be  achieved  by  increasing  the  frequency  of longitudinal observations.  Keywords:  longitudinal data,  joint modeling, robust
GENERALIZED VARIANCE FUNCTIONS FOR BINOMIAL VARIABLES IN STRATIFIED TWO-STAGE SAMPLING Ari Handayani; Aunuddin .; Indahwati .
FORUM STATISTIKA DAN KOMPUTASI Vol. 10 No. 1 (2005)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (552.919 KB)

Abstract

      This empirical study evaluates the application of Generalized Variance Functions (GVFs) for binomial variables in the 1998 Indonesian Labor Force Survey. The survey employs stratified two-stage cluster sampling for selecting samples from a population of households. The study covers all provinces in Java to produce estimates at the level of Java Island. The relative variance estimates resulted from the GVF models are compared to the relative variance estimates which are computed directly. The results illustrate that  model  expressed by logarithmic model  log = log c + d log () gives a good approximation to estimate the variances for the nonagricultural employment group, especially for working male category both in urban and rural areas. It is also good for the total employment group differentiated by age group, educational attainment, and employment status. On the other hand, the model gives poor results for the agricultural employment group. Based on the empirical results, the GVF models may not perform particularly well for the common characteristics which have relatively dissimilar deff values to majority of characteristics in the same group, since these characteristics usually come out among all persons in the sample household and often among all households in the sample cluster as well. The success of the GVF technique depends critically on the grouping of the estimates total () and amount of characteristics involved as the observations for fitting the model. Furthermore, observations with relatively large residuals will also determine the performance of goodness-of-fit of the model. Application of GVF technique to obtain an approximate standard error on numerous binomial characteristics in large scale survey should be carried out further using extensive data. The better performance of GVF model may also be accomplished by utilizing, for examples, weighted least squares procedure or robust regression method. Additionally, the data users should be warned that there will inevitably be survey characteristics for which GVF's will give poor results or even no GVF will be appropriate. Keywords :  Generalized Variance Functions, Stratified Two-Stage Sampling
PERFORMANCE COMPARISON BETWEEN KIMURA 2-PARAMETERS AND JUKES-CANTOR MODEL IN CONSTRUCTING PHYLOGENETIC TREE OF NEIGHBOUR JOINING Hendra Prasetya; Asep Saefuddin; Muladno .
FORUM STATISTIKA DAN KOMPUTASI Vol. 16 No. 1 (2011)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (307.91 KB)

Abstract

Bioinformatics as a recent improvement of knowledge has made an interest for scientist to collect and analyze data to provide the best estimate of the true phylogeny. The objective of this research is to construct and compare the phylogenetic tree of Neighbour Joining (NJ) based on different models (Kimura 2-Parameters and Jukes-Cantor) and to find out which model is more reliable on constructing NJ's tree. In order to build the tree, reliable set of data is conducted from D-loop mtDNA sequences that is available in Gen Bank. The nucleotide sequences come from Bison bison (American bison), Bos taurus (European cow such as Shorthorn), Bos indicus (zebu breeds), Bos grunniens mutus (one of subspecies of cow), and Capra hircus (species of goat). The reliability of each models was measured using the Felsentein's bootstrap method. The whole bootstrap process for each models was repeated 1.000, 5.000, and 10.000 times to detect its reliability. The performance was measured on the basis of the consistency of the topology relationship, the stability of nodes, the consistency of bootstrap confidence level (PB), standard error of distance, change of PB from (1.000-5.000) to (5.000-1.000), computational time, and  BIC score. NJ's phylogenetic tree with kimura 2-parameters and jukes cantor model have a good node stability and is also generally successful in representing topological relationships between taxa. The increasing of bootstrap replication number in common will increase the consistency of bootstrap confidence value ( . It means both models have a good reliability. But, when the number of sequences is large and the extent of sequence divergence is low, it is generally difficult to construct the tree by any models. In conclusion, Kimura 2-Parameters has a better performance than Jukes-Cantor.   Key words: phylogenetic tree, Neighbour Joining, Kimura 2-Parameters, Jukes-Cantor

Page 4 of 12 | Total Record : 119

 2   3   4   5   6 

Filter by Year

2001 2015


Reset
Filter By Issues
All Issue Vol. 20 No. 2 (2015) Vol. 18 No. 1 (2013) Vol. 17 No. 2 (2012) Vol. 17 No. 1 (2012) Vol. 16 No. 2 (2011) Vol. 16 No. 1 (2011) Vol. 15 No. 2 (2010) Vol. 15 No. 1 (2010) Vol. 14 No. 2 (2009) Vol. 14 No. 1 (2009) Vol. 13 No. 1 (2008) Vol. 12 No. 2 (2007) Vol. 12 No. 1 (2007) Vol. 11 No. 2 (2006) Vol. 11 No. 1 (2006) Vol. 10 No. 2 (2005) Vol. 10 No. 1 (2005) Vol. 9 No. 2 (2004) Vol. 8 No. 2 (2003) Vol. 8 No. 1 (2003) Vol. 7 No. 2 (2002) Vol. 6 No. 2 (2001) Vol. 6 No. 1 (2001) More Issue