Indonesian Journal of Statistics and Its Applications
Indonesian Journal of Statistics and Its Applications (eISSN:2599-0802): diterbitkan berkala 2 (dua) kali dalam setahun yang memuat tulisan ilmiah yang berhubungan dengan bidang statistika dan aplikasinya. Artikel yang dimuat berupa hasil penelitian bidang statistika dan aplikasinya dengan topik (tapi tidak terbatas): rancangan dan analisis percobaan, metodologi survey dan analisis, riset operasi, data mining, pemodelan statistika, komputasi statistika, time series dan ekonometrika, serta pendidikan statistika.
Articles
16
Articles
​
BINOMIAL REGRESSION IN SMALL AREA ESTIMATION METHOD FOR ESTIMATE PROPORTION OF CULTURAL INDICATOR

Yudistira, Yudistira, Kurnia, Anang, Soleh, Agus Mohamad

Indonesian Journal of Statistics and Applications Vol 2 No 2 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar |

Abstract

In sampling survey, it was necessary to have sufficient sample size in order to get accurate direct estimator about parameter, but there are many difficulties to fulfill them in practice. Small Area Estimation (SAE) is one of alternative methods to estimate parameter when sample size is not adequate. This method has been widely applied in such variation of model and many fields of research. Our research mainly focused on study how SAE method with binomial regression model is applied to obtained estimate proportion of cultural indicator, especially to estimate proportion of people who appreciate heritages and museums in each regency/city level in West Java Province. Data analysis approach used in our research with resurrected data and variables in order to be compared with previous research. The result later showed that binomial regression model could be used to estimate proportion of cultural indicator in Regency/City in Indonesia with better result than direct estimation method.

PENGGEROMBOLAN DESA/KELURAHAN BERDASARKAN INDIKATOR KEMISKINAN DENGAN MENERAPKAN ALGORITMA TSC DAN K-PROTOTYPES

Munthe, Andrew Donda, Sumertajaya, I Made, Syafitri, Utami Dyah

Indonesian Journal of Statistics and Applications Vol 2 No 2 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar |

Abstract

Statistic Indonesia (BPS) noted that in 2014 there were 3.270 villages in Nusa Tenggara Timur Province. Most of them have a high percentage of poverty. Therefore, the village clustering based on poverty indicators is very important. The clustering algorithm that can be used on large data size and with mixed variables are Two Step Cluster (TSC) and K-Prototypes. The purpose of this research is to compare of TSC and K-Prototypes algorithm for village clustering in Nusa Tenggara Timur Province based on poverty indicators. The data were taken from 2014 village potential data (PODES 2014) collected by BPS. The best selection criteria for the cluster is the minimum ratio between variance within groups and variance between groups. The result showed that the best clustering algorithm was TSC which had the smallest ratio (2.6963). The best clustering showed that villages in Nusa Tenggara Timur Province divided into six groups with different characteristics.

MODELLING THE NUMBER OF NEW PULMONARY TUBERCULOSIS CASES WITH GEOGRAPHICALLY WEIGHTED NEGATIVE BINOMIAL REGRESSION METHOD

Mumtaz, Tsuraya, Utomo, Agung Priyo

Indonesian Journal of Statistics and Applications Vol 2 No 2 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar |

Abstract

Tuberculosis (TB) is an infectious disease caused by Mycobacterium Tuberculosis. Untill now, TB is still one of the main problems in many countries, especially developing countries. Indonesia ranked second as the country with the highest TB cases in the world in 2015, where most cases were found in Java. This study was conducted to model the number of new pulmonary TB cases in Java by considering the spatial aspects using Geographically Weighted Negative Binomial Regression (GWNBR). GWNBR method was chosen  because the data used in this study are overdispered. The result showed that the population density and percentage of healty homes were not significantly influential in each region. While the number of puskesmas, the percentage of smokers, the percentage of good PHBS, the percentage of diabetes mellitus, and the percentage of less IMT were significant in some regions. In general, the GWNBR model was better for modelling the number of new pulmonary TB cases than negative binomial regression and GWPR.

ESTIMASI KEBUTUHAN IMPOR DAGING SAPI UNTUK KONSUMSI RUMAH TANGGA DI INDONESIA MENGGUNAKAN REGRESI ROBUST

Ratnasari, Ratnasari, Sastri, Ray

Indonesian Journal of Statistics and Applications Vol 2 No 2 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar |

Abstract

Beef import to Indonesia always gets pros and cons. The government argue that we need it to reduce the high price of beef due to the scarcity. On the other hand, Indonesia is an agrarian country with a lot of cattle farms. We should be able to meet the needs of beef from domestic production without import. The aim of this study is to get the best model for household consumption of beef at the district level, and use the model to estimate the import needs. This study uses data from Statistics Indonesia, both the raw data of National Sosio-economic Survey (SUSENAS) and beef production in district level. The methods of analysis is a robust regression model. The results is robust regression fit the data well. For households need, estimation of household consumption of beef is lower than domestic production. So that, Indonesia does not need to import beef for household need.

GEOGRAPHICALLY WEIGHTED LOGISTIC REGRESSION DENGAN FUNGSI KERNEL FIXED GAUSSIAN PADA KEMISKINAN JAWA TENGAH

Wulandari, Wulandari

Indonesian Journal of Statistics and Applications Vol 2 No 2 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar |

Abstract

Poverty alleviation is a problem faced by many countries in the world, included Indonesia. Poverty in Indonesia still relatively high. Poverty is one indicator of welfare. In general, the decline in poverty means that people's welfare increasing. Poverty is a multi-dimensional problem, which involves various microeconomic and macroeconomic factors, including the influence of the surrounding region. Modeling with geographically weighted regression (GWR) accommodates heterogeneous effects of independent variables on the dependent variable and produces a local parameter estimates. Central Java has the second highest poverty rate among provinces in Java. This study will model poverty in Central Java with a model that accommodates the influence of the surrounding region, named Geographically Weighted Logistic Regression (GWLR). Poverty modeling in Central Java with GWLR, in general, literacy rates (AMH), per capita GRDP, and Labor Force Participation Rate (TPAK) significantly affected poverty in Central Java with values that varied between districts / cities.

ANALISIS AMMI DENGAN RESPON GABUNGAN PADA UJI STABILITAS TANAMAN PADI GOGO DI KABUPATEN PACITAN

Fahmi, Abdullah Ilman, Anisa, Rahma, Kurnia, Anang

Indonesian Journal of Statistics and Applications Articles in Press
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar |

Abstract

Gogo rice is one of the results of various rice cultivation development by planting in a dry land. Gogo rice is expected to give yield a better production of paddy in dry rice fields. The varieties Inpago 7, Inpago 8, Inpago 8 IPB, Inpago 9, Inpago 10, Situ Gintung, Situ Patenggang, Situ Bagendit, Gajah Mungkur, Slengreng TG, Slegreng GK, Srijaya, Towuti, Merah Wangi, dan Inpari 24 were used in this study. This study aims to identify the Gogo rice varieties that are stable and superior in six Pacitan Garden Experimental Plant locations based on a combined response using the AMMI method. The AMMI analysis combines an additive variety analysis as the main effects of treatment with multiple principle component analysis by bilinier modeling for interaction effect. This study used two combined responses, which described the plant productivity and the resistancy. The result of this study explained that some varieties, Inpago 8, Inpago 10, and Situ Patenggang, were stable varieties in all planting location based on the combined responses. According to productivity stability and plant resistancy superior gogo rice variety is Inpago 8 and Inpago 10.

POISSON REGRESSION OF DAMAGE PRODUCT SALES USING MCMC

Marliana, Reny Rian, Padmadisastra, Septiadi

Indonesian Journal of Statistics and Applications Vol 2 No 1 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar | Full PDF (1038.238 KB)

Abstract

In this paper a model for the number of “damage” product sales is studied. The product sales are run into underreporting counts, caused by a delay on input process of the system called sales cycle. The goal of the study is to estimate the parameters of the regression model of product sales on an explanatory variable. It is the actual number of product sales. The model used is a mixture of the Poisson and the Binomial distributions. The parameters of the regression model are estimated by a Bayesian approach and Markov Chain Monte Carlo simulation using Gibbs sampling algorithm. The results of estimation clearly showed a gap between undamage product sales and the actual number. The gap is the number of damaged product sales.

ALTERNATIF PENGGEROMBOLAN DATA DERET WAKTU DENGAN KONDISI TERDAPAT DATA KOSONG

Yanti, Yusma, Rahardiantoro, Septian

Indonesian Journal of Statistics and Applications Vol 2 No 1 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar | Full PDF (600.849 KB)

Abstract

Panel data describes a condition in which there are many observations with each observation observed periodically over a period of time. The observation clustering context based on this data is known as Clustering of Time Series Data. Many methods are developed based on fluctuating time series data conditions. However, missing data causes problems in this analysis. Missing data is the unavailability of data value on an observation because there is no information related to it. This study attempts to provide an alternative method of clustering observations on data with time series containing missing data by utilizing correlation matrices converted into Euclid distance matrices which are subsequently applied by the hierarchical clustering method. The simulation process was done to see the goodness of alternative method with common method used in data with 0%, 10%, 20% and 40% missing data condition. The result was obtained that the accuracy of the observation bundling on the proposed alternative method is always better than the commonly used method. Furthermore, the implementation was done on the annual gini ratio data of each province in Indonesia in 2007 to 2017 which contained missing data in North Kalimantan Province. There were 2 clusters of province with different characteristics.

FAKTOR-FAKTOR YANG MEMENGARUHI UNMET NEED KB DI PROVINSI BENGKULU TAHUN 2015 DENGAN PEMODELAN REGRESI LOGISTIK BINER

Rai, Abyan, Ramadhan, Reza Rizky

Indonesian Journal of Statistics and Applications Vol 2 No 1 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar | Full PDF (488.424 KB)

Abstract

Indonesia failed to reach the target to reduce the percentage of unmet need KB in 2015.  Province of Bengkulu which has almost 20% of family planning services from government, has a high unmet need percentage. The purpose of this research is to determine the factors that affect status of unmet need KB of women aged 15-49 years in Bengkulu 2015. Data in this study obtained from Susenas 2015. The result showed that 21,34% of women aged 15-49 years in Bengkulu in 2015 was unmet need KB and 78,65% was not unmet need KB. Using binary logistic regression, the result showed that the age of women, number of surviving children, education of women aged 15-49 years, and type of residence have a significant effect on status of unmet need KB. Socialization of family planning program on Bengkulu is needed to reduce the percentage of unmet need KB of women aged 15-49 years on Bengkulu. Further research is suggested to use other independent variables and see spatial correlation in Province of Bengkulu.

REGRESI POISSON BIVARIAT DENGAN KOVARIAN MERUPAKAN FUNGSI DARI VARIABEL BEBAS

Kurniawan, Untung

Indonesian Journal of Statistics and Applications Vol 2 No 1 (2018)
Publisher : Departemen Statistika, IPB dengan Forum Perguruan Tinggi Statistika (FORSTAT)

Show Abstract | Original Source | Check in Google Scholar | Full PDF (560.385 KB)

Abstract

Poisson regression is a regression model which often used to analyze the count data. In this study, poisson regression has been used bivariate poisson regression where the regression is a method which is used to model a pair of correlated count data with multiple predictor variables. The model is used covariance which has a function of the independent variable. The purposes of this study is obtain parameter estimates, test statistics of bivariate poisson regression, and determine the factors that influence of infant mortality and maternal mortality. The data is used from the infant mortality and maternal mortality in Central Java 2015. Based on the result, the parameter estimation of poisson bivariate regression model using maximum likelihood (MLE) method. The results obtained from the parameter estimation are not close form so it needs to be done by Newton-Raphson iteration method. In testing the hypothesis using the Maximum Likelihood Ratio Test method (MLRT) by comparing the value between likelihood below H0 and likelihood below population. Partial of parameters model ?1 (infant mortality) there are six independent variables that have significant influence, namely, delivery by health personnel (X1), pregnant women carry out the program K4 (X3), pregnant women who get Fe3 tablet (X4), handling obstetric complication (X5), exclusively breastfed infants (X7), and households living a clean and healthy life (X8). While for model ?2 (maternal death) only variable handling of neonatal complication (X6) which have no significant influence to response variable.