FORUM STATISTIKA DAN KOMPUTASI
Not a member yet
119 research outputs found
Sort by
PENDEKATAN NONPARAMETRIK UNTUK ANALISIS TREND PADA RESPONS BINER
No full textPada saat penelitian lebih difokuskan pada proporsi dari banyaknya ‘sukses’, pi = Yi/Ni, maka analisis seringkali dilakukan berdasarkan model sampling untuk proporsi:distribusi binomial. Distribusi statistik sederhana seperti binomial kadang-kadang tidak mampu untuk menggambarkan distribusi sampling dari Yi atau pi. Dengan demikian, untuk setiap analisis berdasarkan pada penaksiran parameter dari model binomial (yaitu metode parametrik binomial) akan membawa pada kekeliruan dalam inferensi mengenai efek dari suatu stimulus yang sedang diamati. Dalam makalah ini akan dibahas mengenai suatu alternatif dari model parametrik untuk pi, yaitu dengan menggunakanmetode bebas-distribusi (nonparametrik). Dua buah metode berdasarkan pendekatan nonparametrik untuk keperluan analisis trend yang akan dibahas dalam makalah ini ujiCochran-Armitage dan uji Permutasi
PENERAPAN METODE JACKKNIFE DALAM PENDUGAAN AREA KECIL
No full textMetode dasar yang sering digunakan dalam menyelesaikan model pendugaan tidak langsung pada SAE adalah BLUP/EBLUP, EB, dan HB. Namun ketidakpuasan sering muncul karena asumsi kelinieran atau sebaran tertentu tidak selalu dipenuhi dalam suatu analisis. Selain itu, penambahan komponen g2 dan g3 dari MSE( ?ˆi BP) tidak lain adalah upaya untuk mengkoreksi ketidakpastian akibat terlebih dulu melakukan pendugaan terhadap b dan s2u. Dengan teknik resampling, jackknife berkembang sebagai suatu metode untuk mengkoreksi biassuatu penduga. Penerapan jackknife pada pendugaan area kecil dilakukan untuk mengkoreksi pendugaan MSE
Pengembangan Aplikasi Perangkat Lunak Regresi Komponen Utama
No full textPenangangan kasus multikolinier pada analisis regresi linier ganda dilakukan melalui berbagai metode, salah satunya adalah dengan menggunakan regresi komponen utama. Software-software statistik yang ada saat ini belum memberikan suatu langkah yang mudah dalam melakukan analisis regresi.komponen utama. Dengan menggunakan bahasa pemrograman C++ dikembangkan software SiRegLin, yang merupakan software untuk pendugaan model linier regresi pada kasus terjadinya multikolinier dengan komponen utama. SiRegLin dikembangkan pada sistem operasi berbasis Microsoft Windows dengan perangkat lunak pengembangan menggunakan Borland C++ Builder versi 6.0. Kata Kunci : Multikolinearitas, Regresi Komponen Utam
PERBANDINGAN KUASA UJI PENDEKATAN BIGGERS DAN SATTERTHWAITE-COCHRAN DALAM MENGANALISIS DATA HILANG PADA RANCANGAN KELOMPOK TERACAK LENGKAP
Get PDFThis study was aimed to compare type I error in estimating missing data by Biggers and Satterhwaite-Cochran methods in Randomized Comple Block Design. Simulation study using Gauss software showed that type I error for Biggers approach are over estimate. The problem can be overcomed by Satterhwaite-Cochran approach.Keyword: Missing data, randomized Block design, Biggers, Satterhwaite-Cochran
KLASIFIKASI GENOTIPE PADA DATA TIDAK LENGKAP DENGAN PENDEKATAN MODEL AMMI
Get PDFPercobaan multilokasi mempunyai peranan penting dalam perkembangbiakan tanaman dan penelitian agronomi. Kajian mengenai interaksi antara genotipe dan lingkungan diperlukan dalam penyeleksian genotipe yang akan dilepas. Metode statistika yang biasa digunakan untuk mengolah data hasil percobaan multilokasi salah satunya adalah AMMI (Additive Main effect and Multiplicative Interaction). Metode ini menggabungkan analisis ragam aditif bagi pengaruh utama perlakuan dengan analisis komponen utama pada pengaruh interaksinya. Pendekatan AMMI juga sangat baik digunakan untuk uji multilokasi tanpa ulangan. AMMI adalah analisis yang membutuhkan data yang lengkap. Jika ada data yang hilang, maka harus dilakukan pendugaan terhadap data tersebut. Pada kasus data tidak lengkap, diperlukan suatu metode pendugaan data untuk mempermudah analisis. Metode yang dapat digunakan antara lain connected data dan algoritma EM-AMMI untuk menduga data yang tak lengkap
PENGGUNAAN ALGORITMA SIMULATED ANNEALING UNTUK MENYELESAIKAN TEKA-TEKI BINARY DAN SUDOKU ( Solving Binary and Sudoku Puzzles with a Simulated Annealing Algorithm )
Get PDFBinary 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 annealin
PENGARUH PEMILIHAN ARAH ACUAN 00 DAN ARAH ROTASI PADA ANALISIS KORELASI DAN REGRESI LINIER-SIRKULAR (STUDI KASUS: PETA KAWASAN RAWAN BENCANA LETUSAN GUNUNG
Get PDFThe measurement results doesn\u27t only consist of data with linear attributes, but also data with circular attributes. The circular data has a uniqueness that is not owned by the linear data, circular data is independent of the choice of 0o reference and rotation direction. The uniqueness of circular data analysis is tested in linear circular correlation and linear circular regression. The results of correlation analysis proved that the selection of the reference direction 0o can be done subjectively because the linear circular correlation results show the same value 0.899 for all possible selection of 0o reference and rotation direction. For linear circular regression, the model constructed has a same coefficient of determination that is 0.808 and the same b0, which is 5.231 for all possible selection of 0o reference and rotation direction. Similarly, statistics from the error of linear circular regression analysis have the same value, minimum = -2.693, quartile 1 = -0.835, median = -0.171, quartile 3 = 0.548, maximum = 8.421. Alleged circular linear regression parameters, namely b1 and b2, forming a cycle that each has in common b1 = -1.226 E-07-2.728 cos (α) - 2.655 sin (α) and b2 = 3.061 E-07-2.655 cos (α ) + 2.728 sin (α) where α is the position of the 0o reference direction in degrees on each model. Keywords : Directional Statistics, Circular Statistics, Linear-Circular Regression, Linear Circular Correlatio
JARINGAN SYARAF TIRUAN DAN ALGORITMA GENETIKA DALAM PEMODELAN KALIBRASI (STUDI KASUS : TANAMAN OBAT TEMULAWAK)
Get PDFThe problems in prediction of calibration model are multicolinearity and the number of variables is larger than the number of observations. Principal Component Analysis-Artificial Neural Network-Genetic Algorithm (PCA-ANN-GA) models were applied for the relationship between sample of concentration which is limited and transmittance data which is in large dimensions. A large number of variables were compressed into principal components (PC’s). From these PC’s, the ANN was employed for prediction of concentration. The principal components computed by PCA were applied as inputs to a backpropagation neural network with one hidden layer. The models was evaluated using GA for the best network structure on hidden layer. Root Mean Square Error (RMSE) for 80% training set and 20% testing set are 0.0314 and 0.5225, respectively. Distribution of data according to the percentage of training data and testing data were also very influential to obtain the best network structure with the smallest RMSE achievement. The best model for these methods is two layers Neural Network with eight neuron in the hidden layer
PENDETEKSIAN PERILAKU HERDING PADA PASAR SAHAM INDONESIA DAN ASIA PASIFIK (Detection of Herding Behavior on Indonesia and Asia Pacific Stock Market)
Get PDFHerding 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, CSA
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)
Get PDFJoint 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, robus