Garuda - Garba Rujukan Digital

PEMILIHAN PENGRAJIN TERBAIK MENGGUNAKAN MULTI-ATTRIBUTE DECISION MAKING (MADM) TECHNIQUE FOR ORDER PREFERENCE BY SIMILARITY TO IDEAL SOLUTION (TOPSIS) (STUDI KASUS : PT. Sinjaraga Santika Sport, Majalengka) Fizry Listiyani Maulida; Tatik Widiharih; Alan Prahutama
Jurnal Gaussian Vol 4, No 2 (2015): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

The human resources (HR) known as the employess are the successful of the company. PT. Sinjaraga Santika Sport (Triple’S) is a handmade football company by the craftsmen. Most of the craftsmen go to the rice fields on the growing season or the harvest season. So selection of the best craftsmen is needed in order to the production of the football don’t have problems. The selection uses TOPSIS method. TOPSIS is one of method that can be used to solve MADM problem. The steps of TOPSIS method are calculated the normalized decision matrix, determined the weight, calculated the weighted normalized decision matrix, determined the positif-ideal solutions and negatif-ideal solutions, calculated the separation measures, and calculated the preference value. There are 25 craftsmen and six criteria. The criteria are neatness of the ball, accurateness stitching of the ball, number of the ball, accurateness logo of the ball, cleanness of the ball, and defect proportion. The results in this reseach are the best carftsmen has 0,78861 of preference value and the worst craftsmen has 0,16642 of preference value. Preference value by manual calculate equal with preference value by GUI Matlab. Keywords : TOPSIS, MADM, carftsmen

RANCANGAN D-OPTIMAL LOKAL UNTUK REGRESI POLINOMIAL ORDE 3 DENGAN HETEROSKEDASTISITAS Arya Fendha Ibnu Shina; Tatik Widiharih; Triastuti Wuryandari
Jurnal Gaussian Vol 1, No 1 (2012): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Kemajuan ilmu pengetahuan dan teknologi di berbagai bidang menuntut adanya rancangan percobaan yang efisien. Rancangan D-optimal merupakan rancangan yang efisien. Dalam suatu percobaan yang menggunakan model regresi polinomial orde dengan heteroskedastisitas dengan fungsi bobot , rancangan D-optimal dan polinomial Jacobi menghasilkan titik-titik rancangan yang akan dicobakan. Suatu rancangan yang terdiri dari titik-titik rancangan dengan proporsi pengamatan yang menghasilkan determinan matriks rancangan maksimal merupakan rancangan D-Optimal. Rancangan D-optimal yang memiliki nilai variansi terstandardisasi sama dengan jumlah parameter di setiap titiknya, merupakan rancangan D-optimal lokal.

PENDEKATAN SISTEM PERSAMAAN SIMULTAN DALAM PEMODELAN PRODUK DOMESTIK REGIONAL BRUTO (PDRB) PROPINSI JAWA TENGAH TAHUN 2000-2010 Rizky Oky Ari Satrio; Tatik Widiharih; Abdul Hoyyi
Jurnal Gaussian Vol 1, No 1 (2012): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Gross Domestic Product (GDP) is general indicator used to identify the economical development in a region. The condition of economy in Central Java Province is categorized as stable condition since it has GDP value developed rapidly year by year. Refer to model used by Bappenas,the simultaneous equation model between GDP is influenced by number of employee and government spending.Identification of the model in this study using the ordercondition of indetification on the basis of the result of the overidentified taken the GDP of agriculture, mining, electricity, gas and water sector and trade. Therefore, the parameter evaluation used is 2SLS method (Two Stage Least Square). After fulfilled assumption of independent, identical and normal distribution, the most valued toward model of GDP in Central Java Province is GDP sector of agriculture.

APLIKASI MODEL REGRESI POISSON TERGENERALISASI PADA KASUS ANGKA KEMATIAN BAYI DI JAWA TENGAH TAHUN 2007 Nurwihda Safrida Umami; Dwi Ispriyanti; Tatik Widiharih
Jurnal Gaussian Vol 2, No 4 (2013): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Infant Mortality is one of the issues that can affect the number and age composition of the population. The Government pays special attention to reduce the amount of Infant Mortality Rate in Central Java, so the role of data and information becomes very important. Poisson regression is a nonlinear regression which is often used to model the relationship between the response variable in the form of discrete data with predictor variables in the form of continuous or discrete data. Poisson regression models have equidispersi assumption, a condition in which the mean and variance of the response variable have equal value. In practice, the assumption is sometimes violated in the analysis of discrete data in the form of overdispersi (value of variance greater than the mean value) so that Poisson regression model is not appropriate to be used. Overdispersi is a condition in which the data of response variable shows. One model that can be used to solve the overdispersi problem is generalized Poisson regression model. The regression model is an extension of the Poisson regression and part of the Generalized Linear Model (GLM) which does not require constancy of variance to test the hypothesis. From the data of Infant Mortality Rate in Central Java on 2007 known that there overdispersi. And the factors affecting Infant Mortality Rate is the number of health facilities, the number of medical personnel, and the percentage of households with clean water each county / city.

PEMODELAN PENDAPATAN ASLI DAERAH (PAD) DI KABUPATEN DAN KOTA DI JAWA TENGAH MENGGUNAKAN GEOGRAPHICALLY WEIGHTED RIDGE REGRESSION Depy Veronica; Hasbi Yasin; Tatik Widiharih
Jurnal Gaussian Vol 5, No 3 (2016): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Linear Regression Analysis is a statistical method for modeling the relation between response variable and predictor variable. Geographically Weighted Regression (GWR) is an expansion of linear regression model if spatial heterogeneity occurred. Local multicollinearity test is required to know the presence of linear correlation between independent variables for each observation location. Geographically Weighted Ridge Regression (GWRR) is a extension of GWR model to solve local multicollinearity problem. Parameter estimation for GWR and GWRR model is done using Weighted Least Square (WLS) method by applying optimum bandwith with Cross Validation (CV) criteria. GWRR model is applied on locally generated recurring revenues (PAD) at district and city of Central Java and its result shows the ability of GWRR model to erase multicollinearity problem. Based on Mean Squared Error (MSE) and Akaike Information Criterion (AIC) value for GWR and GWRR model, it is know that the best model to analyze locally generated recurring revenues (PAD) at district and city of Central Java is GWRR model with the smallest MSE and AIC value. Keywords : Akaike Information Crietion, Spasial Heterogeneity, Geographically Weighted Ridge Regression, Mean Square Error, Local Multicoliniearity

PREDIKSI RETURN PORTOFOLIO MENGGUNAKAN METODE KALMAN FILTER Dita Rosita Sari; Tatik Widiharih; Sugito Sugito
Jurnal Gaussian Vol 5, No 4 (2016): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Stock is an evidence for individual or institutional ownership about a company. To cover losses in stocks investment, should be done diversification to spread risk in some stocks called as portfolio. Portfolio is a joint of two or more stocks investment that are choosen as investment’s targets over spesific time periods and certain rules. To minimize losses in stocks investment, needed to predict portfolio return for some coming periods. Good prediction has small difference with actual data. One method that can minimize MSE is Kalman Filter. Kalman Filter estimates a process through feed back Control Mechanism called recursion. The variable used are monthly portfolio return of PT Mayora Indah Tbk and PT Indofood Sukses Makmur Tbk in January 2005 until December 2015. Data In January 2005 until December 2014 are used to predict the return portfolio for Year 2015. After that, an interval is made for those forecast results and compare with actual data. If actual data are residing in the interval, then Kalman Filter method can be used to predict portfolio return for year 2016. The MSE value with kalman Filter is 0,00225 and the MSE value with Box-Jenkis method is 0,00253, so Kalman Filter can minimize the MSE value. Keywords : portfolio return, Box-Jenkins, Kalman Filter

RANCANGAN ACAK KELOMPOK TAK LENGKAP SEIMBANG PARSIAL (RAKTLSP) Gustriza Erda; Tatik Widiharih; Yuciana Wilandari
Jurnal Gaussian Vol 4, No 2 (2015): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Partially Balanced Incomplete Block Designs (PBIBD) is a design with v treatments arranged into b blocks with every block which is consist of into k treatment (k < v) that in every treatment only occurs once in every block, and there are pair treatment which occur together in the same block as much as λm times. The pair treatments on PBIBD is based on the association scheme. This undegraduate thesis uses triangular association scheme that is two-class association scheme (first and second association). This scheme is used to determine the first and second association of every treatment. Based on formed association, it will obtain the number of pairs treatment that occurs in every block that will be designed (λm, m=1,2). The test that is used is test of treatments effect because only treatments that is important which are adjusted treatment for the reason that not all treatments occurs in every block. Assumptions which is required is the assumption of residual normality, equal variances, and independence assumption. The advanced test to be held is Tuckey Test (Honest Significance Difference). To clarify the discussion on PBID, examples of applications in the field of animal husbandry are given to observe the effect of the type of foods that contain alfalfa effect toward weight gain of turkey. The result obtained indicate that there are significant types of foods that contain alfalfa effect toward weight gain of turkey. Where is the recommended type of food is the food of A that contain 2,5% alfafa type 22.Keywords : PBIBD, Triangular association, Tuckey Test, Normality, Equal Variances, Independence

PEMILIHAN HELM TERFAVORIT DENGAN MADM BERBASIS GUI MATLAB (Studi Kasus : Pemilihan Helm Terfavorit oleh Mahasiswa FSM Undip, Semarang) Nadya Kiki Aulia; Tatik Widiharih; Abdul Hoyyi
Jurnal Gaussian Vol 6, No 3 (2017): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Safety is an important factor that need to be considered in driving safety. One of important factor that need to be considered is the use of Indonesian National Standard (SNI) helmets. The large number of SNI helmets existance, make consumers especially students, have their own preferences in choosing a helmet. The methods that can be used to choose the most favorite helmet is ELECTRE and TOPSIS. Both of these methods are the methods used to solve MADM problems. There are 8 brands of helmets namely INK, KYT, BMC, HIU, VOG, BOGO, NHK, dan GM. For helmet selection criteria are safety helmets (helmet safety straps when worn), affordable helmet prices, easy found helmet, variety of helmet colors, various sizes available, helmets cover the entire face, comfortable helmet glass when worn, clear helmet glass, quality of the outside of the helmets, helmet foam quality, and resistance to impact. By using ELECTRE method, this research got result that the most favorite helm is INK helmet brand which has the number of row element as much 5. For TOPSIS method, the most favorite helmet is KYT helmet brand with preference value equal to 0.7146. Keywords: ELECTRE, TOPSIS, Helmet, favorite, GUI

RANCANGAN D-OPTIMAL UNTUK REGRESI POLINOMIAL DUA FAKTOR DERAJAT DUA Rosmalia Safitri; Tatik Widiharih; Triastuti Wuryandari
Jurnal Gaussian Vol 1, No 1 (2012): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Suatu penelitian dalam bidang kimia seringkali memerlukan suatu rancangan optimal untuk menentukan titik mana dari variabel prediktor yang akan dicobakan dengan tujuan memaksimalkan sejumlah informasi yang relevan sehingga terpenuhi kriteria yang diinginkan. Kriteria pemenuhan optimal didasarkan pada matriks rancangan dari model yang dipilih. Kriteria D-optimal digunakan untuk meminimalkan variansi dari estimasi parameter dengan cara memaksimalkan determinan matriks informasinya atau meminimalkan determinan matriks dispersinya. Pemilihan titik-titik dari variabel prediktor selain tergantung dari model yang dipilih juga tergantung dari banyaknya pengamatan yang diinginkan.Kriteria D-optimal diaplikasikan pada data simulasi untuk kasus pengukuran nilai persentase kelarutan enam reaksi kimia berdasarkan nilai suhu dan lama reaksinya. Diperoleh kesimpulan bahwa determinan matriks informasi maksimal terjadi pada saat iterasi keempat dengan nilainya sebesar 2.2070 x 109.

ANALISIS DESAIN FAKTORIAL FRAKSIONAL 2k-p DENGAN METODE LENTH Gian Kusuma Diah Tantri; Tatik Widiharih; Triastuti Wuryandari
Jurnal Gaussian Vol 4, No 3 (2015): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Rancangan faktorial fraksional banyak digunakan dalam percobaan terutama di bidang industri karena dapat menentukan pengaruh faktor utama dan interaksi terhadap respon. Rancangan yang melibatkan k buah faktor dengan dua taraf dan menggunakan 2-p fraksi dari percobaan faktorial lengkap disebut rancangan faktorial fraksional 2k-p. Penentuan faktor signifikan jika data yang diamati tanpa pengulangan dapat diuji dengan menggunakan metode Lenth. Penelitian ini bertujuan untuk menentukan penaksir dan statistik uji untuk mendapatkan faktor signifikan dengan metode Lenth, serta menentukan perbedaan dalam penggunaan metode Lenth dengan metode klasik. Kasus yang digunakan adalah rancangan faktorial fraksional 25-1 dengan faktor A, B, C, D, E. Hasil pengujian dengan metode Lenth diperoleh nilai estimasi S0 dan sebagai penaksir awal dan akhir. Nilai Margin Error dan Simultan Margin Error sebagai batas kesalahan dalam penentuan faktor signifikan. Faktor yang berpengaruh terhadap respon adalah faktor B dan C. Apabila diuji dengan metode klasik diperoleh faktor yang berpengaruh terhadap respon adalah faktor B, C, D, E, AB, AC, dan BC, sehingga dapat dikatakan bahwa metode klasik lebih sensitif daripada metode Lenth. Kata kunci: Faktorial, fraksional, tanpa pengulangan, plot probabilitas normal, metode Lenth

Title

Found 15 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 15 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 15 Documents
Search