Claim Missing Document
Check
Articles

Found 16 Documents
Search

PEMODELAN VECTOR AUTOREGRESSIVE X (VARX) UNTUK MERAMALKAN JUMLAH UANG BEREDAR DI INDONESIA Rosyidah, Haniatur; Rahmawati, Rita; Prahutama, Alan
Jurnal Gaussian Vol 6, No 3 (2017): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (527.466 KB) | DOI: 10.14710/j.gauss.v6i3.19306

Abstract

The economic stability of a country can be seen from the value of inflation. The money supply in a country will affect the value of inflation, so it is necessary to control the money supply. The money supply in Indonesia consists of currency, quasi money, and securities other than shares. One of the factors affecting the amount of currency, quasi money, and securities other than shares is the SBI interest rate. Time series data from the money supply components are correlated. To explain multiple time series data variables that are correlated we can use the VAR approach. VAR model with the addition of an exogenous variable is called VARX. The purpose of this study is to obtain models to predict the amount of currency, quasi money, securities other than shares using the VARX approach with the SBI interest rate as an exogenous variable. The results of data analysis in this study, the model obtained is VARX (1,1). Based on t test with 5% significance level, SBI interest rate variable has no significant effect to variable of currency amount, amount of quasi money, or amount of securities other than shares. Residual model VARX (1,1) satisfies the white noise assumption, while the normal multivariate assumption is not satisfied. The value of MAPE for currency variables (7,53969%), quasi money (0,49036%), and securities other than shares (9,64245%) indicates that the VARX (1,1) model has excellent forecasting ability that can be used for forecasting future periods. Forecasting results indicate an increase in the amount of currency, quasi money, or securities other than shares in each period..Keywords : Amount of currency, amount of quasi money, amount of securities other than shares, SBI interest rate, VARX, MAPE
INFERENSI STATISTIK DARI DISTRIBUSI NORMAL DENGAN METODE BAYES UNTUK NON-INFORMATIF PRIOR Prahutama, Alan; Sugito, Sugito; Rusgiyono, Agus
MEDIA STATISTIKA Vol 5, No 2 (2012): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (439.544 KB) | DOI: 10.14710/medstat.5.2.95-104

Abstract

One of the method that can be used in statistical inference is Bayesian method. It combine sample distribution and prior distribution to get a posterior distribution. In this paper, sample distribution used is univariate normal distribution. Prior distribution used is non-informative prior. Determination technique of non-informative prior use Jefrrey’s method  from univariate normal distribution. After got the posterior distribution, find the  marginal distribution of mean and variance. So that will get the parameter estimation of interval for mean and variance. Hypothesis testing for mean and variance can find from parameter estimation of formed interval.   Keywords: Bayesian method, non-informatif prior, Jeffrey’s method, Parameter Estimation of Interval, Hypothesis test
APLIKASI GENERALIZED SPACE TIME AUTOREGRESSIVE (GSTAR) PADA PEMODELAN VOLUME KENDARAAN MASUK TOL SEMARANG Anggraeni, Dian; Prahutama, Alan; Andari, Shofi
MEDIA STATISTIKA Vol 6, No 2 (2013): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (625.794 KB) | DOI: 10.14710/medstat.6.2.61-70

Abstract

Time series data from neighboring separated location often associated both spatially and through time. Generalized space time autoregrresive (GSTAR) model is one of the most common used space-time model to modeling and predicting spatial and time series data. This study applied GSTAR to modeling vehicle volume entering four tollgate (GT) in Semarang City: GT Muktiharjo, GT Gayamsari, GT Tembalang, and GT Manyaran. The data was collected by month from 2003 to 2009. The best model provided by this study is GSTAR (21)-I(1,12) uniformly weighted with the smallest REMSE mean 76834. Key words: GSTAR, Vehicle Volume, Space-Time Model
PEMODELAN REGRESI NONPARAMETRIK MENGGUNAKAN PENDEKATAN POLINOMIAL LOKAL PADA BEBAN LISTRIK DI KOTA SEMARANG Suparti, Suparti; Prahutama, Alan
MEDIA STATISTIKA Vol 9, No 2 (2016): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (150.91 KB) | DOI: 10.14710/medstat.9.2.85-93

Abstract

Semarang is the provincial capital of Central Java, with infrastructure and economic’s growth was high. The phenomenon of power outages that occurred in Semarang, certainly disrupted economic development in Semarang. Large electrical energy consumed by industrial-scale consumers and households in the San Francisco area, monitored or recorded automatically and presented into a historical data load power consumption. Therefore, this study modeling the load power consumption at a time when not influenced by the use of electrical load (t-1)-th. Modeling using nonparametric regression approach with Local polynomial. In this study, the kernel used is a Gaussian kernel. In local polynomial modeling, determined optimum bandwidth. One of the optimum bandwidth determination using the Generalized Cross Validation (GCV). GCV values obtained amounted to 1425.726 with a minimum bandwidth of 394. Modelling generate local polynomial of order 2 with MSE value of 1408.672. Keywords: electrical load, local polinomial, gaussian kernel, GCV.
PEMODELAN REGRESI BERGANDA DAN GEOGRAPHICALLY WEIGHTED REGRESSION PADA TINGKAT PENGANGGURAN TERBUKA DI JAWA TENGAH Utami, Tiani Wahyu; Rohman, Abdul; Prahutama, Alan
MEDIA STATISTIKA Vol 9, No 2 (2016): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (303.285 KB) | DOI: 10.14710/medstat.9.2.133-147

Abstract

The problems in employment was the growing number of Open Unemployment Rate (OUR). The open unemployment rate is a number that indicates the number of unemployed to the 100 residents are included in the labor force. The purpose of this study is mapping the data OUR in Central Java and the suspect and identify linkages between factors that cause OUR in the District / City of Central Java in 2014. Factors that allegedly include population density (X1), Inflation (X2), the GDP value (X3), UMR Value (X4), the percentage of GDP growth rate (X5), Hope of the old school (X6), the percentage of the labor force by age (X7) and the percentage of employment (X8). Geographically Weighted Regression (GWR) is a method for modeling the response of the predictor variables, by including elements of the area (spatial) into the point-based model. This research resulted in the conclusion that the OLS regression models have poor performance because the residual variance is not homogeneous. There were no significant differences between GWR models with OLS model or in other words generally predictor variables did not affect the response variable (rate of unemployment in Central Java) spatially. However, GWR model could captured modelling in each region. Keywords: multiple linear regression, geographiically weighted regression, open unemployement rate in Central Java.
PREDIKSI HARGA SAHAM MENGGUNAKAN SUPPORT VECTOR REGRESSION DENGAN ALGORITMA GRID SEARCH Yasin, Hasbi; Prahutama, Alan; Utami, Tiani Wahyu
MEDIA STATISTIKA Vol 7, No 1 (2014): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (335.209 KB) | DOI: 10.14710/medstat.7.1.29-35

Abstract

The stock market has become a popular investment channel in recent years because of the low return rates of other investment. The stock price prediction is in the interest of both private and institution investors. Accurate forecasting of stock prices is an appealing yet difficult activity in the business world. Therefore, stock prices forecasting is regarded as one of the most challenging topics in business. The forecasting techniques used in the literature can be classified into two categories: linear models and non linear models.  One of forecasting techniques in nonlinear models is support vector regression (SVR). Basically, SVR adopts the structural risk minimization principle to estimate a function by minimizing an upper bound of the generalization. The optimal parameters of SVR can be use Grid Search Algorithm method. Concept of this method is using cross validation (CV). In this paper, the SVR model use linear kernel function. The accurate prediction of stock price, in telecommunication, is 92.47% for training data and 83.39% for testing data.   Keywords: Stock price, SVR, Grid Search, Linear kernel function.
ANALISIS DISKRIMINAN BERGANDA DENGAN PEUBAH BEBAS CAMPURAN KATEGORIK DAN KONTINU PADA KLASIFIKASI INDEKS PRESTASI KUMULATIF MAHASISWA Walidaini, Nur; Mukid, Moch. Abdul; Prahutama, Alan; Rusgiyono, Agus
MEDIA STATISTIKA Vol 10, No 2 (2017): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (338.333 KB) | DOI: 10.14710/medstat.10.2.71-83

Abstract

Multiple discriminant analysis is one of the discriminant analysis techniques where the dependent variable  are grouped into more than two groups. This paper discussed how to categorize Grade Point Average (GPA) of undergraduate student of Faculty of Sciences and Mathematics Diponegoro University based on categorical and continuous independent variable including gender, internet usage, time per week for learning, average score in national examination, amount of pocket money per month and the way to enter to Diponegoro University. The GPA grouping refers to the Academic Regulations of Diponegoro University i.e. satisfactory GPA (2,00 to 2,75), very satisfactory (2,76 to 3,50) and with honors (cum laude) (3,51 to 4,00). By using the multiple discriminant analysis with mixture variables, the accuration of classification based on training and testing data reach to 71,875% and 41,667% respectively. 
PEMODELAN BIVARIATE POLINOMIAL LOKAL PADA JUMLAH KEMATIAN IBU DAN BAYI DI JAWA TENGAH Prahutama, Alan; Suparti, Suparti; Ispriyanti, Dwi; Utami, Tiani Wahyu
Prosiding Seminar Nasional Venue Artikulasi-Riset, Inovasi, Resonansi-Teori, dan Aplikasi Statistika (VARIANSI) Vol 1 (2018)
Publisher : Program Studi Statistika, FMIPA, Universitas Negeri Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1082.222 KB)

Abstract

Analisis regresi merupakan analisis dalam metode statistika untuk memodelkan hubungan antara variabel respon dengan variabel prediktor. Analisis regresi dapat dilakukan secara parametrik dan nonparametrik. Analisis regresi nonparametrik dilakukan apabila bentuk kurva regresinya tidak diketahui. Salah satu metode dalam analisis regresi nonparametrik adalah polinomial lokal. Polinomial lokal dilakukan berdasarkan pembobotan kernel, sehingga membutuhkan bandwidth. Pemilihan bandwidth optimal menggunakan Generalized Cross Validation (GCV). Pada penelitian ini dikembangkan model regresi bivariate polinomial lokal pada kasus pemodelan jumlah kematian ibu dan bayi di Jawa Tengah. Variabel prediktor yang digunakan adalah jumlah tenaga kesehatan. Nilai bandwidth optimla yang didapatkan adalah 1. Nilai MSE yang dihasilkan dari model jumlah kematian ibu adalah 1.017741 dan Nilai MSE yang dihasilkan dari model jumlah kematian bayi adalah 1.380833. Keywords: Bivariate, Polinomial Lokal, Jumlah kematian ibu, Jumlah kematian bayi.
ANALISIS DATA INFLASI INDONESIA MENGGUNAKAN METODE FOURIER DAN WAVELET MULTISCALE AUTOREGRESIVE Suparti, Suparti; Santoso, Rukun; Prahutama, Alan; Yasin, Hasbi; Devi, Alvita Rachma
Prosiding Seminar Nasional Venue Artikulasi-Riset, Inovasi, Resonansi-Teori, dan Aplikasi Statistika (VARIANSI) Vol 1 (2018)
Publisher : Program Studi Statistika, FMIPA, Universitas Negeri Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (884.754 KB)

Abstract

Analisis regresi merupakan metode statistika untuk mengetahui hubungan antara variabel prediktor dan variabel respon. Pendekatan regresi dapat dilakukan dengan  pendekatan parametrik dan nonparametrik. Pendekatan parametrik ketat dengan asumsi dan harus dipenuhi untuk mendapatkan model yang baik. Sementara pendekatan nonparametrik tidak ketat dengan asumsi karena metode tersebut didasarkan pada pendekatan kurva yang tidak diketahui bentuknya. Pendekatan nonparametrik dapat dilakukan dengan beberapa pendekatan diantaranya metode Fourier dan Wavelet. Metode Fourier merupakan metode yang didasarkan pada deret cosinus atau sinus. Metode Fourier sangat sesuai untuk data yang mengalami pola berulang atau stasioner. Sedangkan pada pemodelan wavelet tidak hanya terbatas pada data berulang atau stasioner saja, akan tetapi juga mampu memodelkan data yang tidak stasioner. Pada penelitian ini dimodelkan nilai Inflasi di Indonesia dari Januari 2007 sampai Agustus 2017.  Variabel responnya adalah nilai inflasi, sedangkan variabel prediktornya adalah waktu. Metode Fourier dengan K=100 menghasilkan MSE sebesar 0,846216 dan R2 sebesar 80,12%. Model Wavelet menggunakan Multiscale Autoregresive dengan filter Haar, J=4 dan Aj = 2  mempunyai MSE sebesar 0,312 dengan R2  sebesar  96,91%.  Pada model Fourier dengan K=100 diperlukan parameter sebanyak 102 buah sedangkan model wavelet dengan J=4 dan Aj = 2 hanya diperlukan parameter sebanyak 10 buah. Jadi model wavelet sangat efisien dengan kinerja yang lebih bagus dibandingkan dengan model Fourier. Kata Kunci: Inflasi, nonparametrik, Fourier, Wavelet, MSE
PEMODELAN INFLASI BERDASARKAN HARGA-HARGA PANGAN MENGGUNAKAN SPLINE MULTIVARIABEL Prahutama, Alan; Utama, Tiani Wahyu; Caraka, Rezzy Eko; Zumrohtuliyosi, Dede
MEDIA STATISTIKA Vol 7, No 2 (2014): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (292.36 KB) | DOI: 10.14710/medstat.7.2.89-94

Abstract

Inflation is defined as a sustained increase in the general level of price for goods and services. Some of the events that led to inflation in Indonesia is rising fuel prices, rising prices of meat and chili. Inflation has negative impact, because decreased purchasing power.  So that the inflation model is needed. Modeling inflation can be use regression models. The approach can be performed with nonparametric regression, one of method of nonparametric regression is spline method. In this case, use three predictors to modeling inflation using spline multivariable. The predictors are price of rice, price of chicken, and price of chili. Obtained multivariable spline models with R-square of 93.94% with optimal m = 2 (quadratic) for 1 knots. Keywords: Spline Multivariable, GCV, Inflation