FORUM STATISTIKA DAN KOMPUTASI
Not a member yet
119 research outputs found
Sort by
PEMODELAN DATA PANEL SPASIAL DENGAN DIMENSI RUANG DAN WAKTU (Spatial Panel Data Modeling with Space and Time Dimensions)
Full text linkThe modeling of spatial panel data is a method of analysis that include the dimension of space and time. In this analysis, the set of data that is required is a combination of cross sections and time series data, that is, either the data observed in each observation location periodically from time to time. On modeling of panel data, there are three approaches, namely pooled least square model, fixed and random effects model. While on modeling of spatial panel data there are several approaches which is a combination of these three approaches in modeling panel data with spatial autoregression model (SAR) and spatial error model (SEM). This research aims to apply a spatial panel data model analysis to include the dimension of space and time in a model. The data that used in this research is GDP, local revenues, a total population and total regional expenditures of ten districts in Jambi province during the years 2000-2008. The results from spatial panel data analysis obtained that model regression of spatial panel data corresponding to the data is panel data models with fixed effect model and spatial error model. From the results of such analysis can also be seen an increase in R2 compared with panel data analysis.Keywords : the modeling of panel data, the modeling of spatial panel data, SAR, SE
PENGGUNAAN SOCIOGRAM UNTUK MENGIDENTIFIKASI POLA JARINGAN SOSIAL PEMBELAJARAN MANDIRI MAHASISWA (Identification of Social Network of Student’s Independent Learning using Sociogram)
Full text linkThis paper presents a useful tool to help universities to increasing the level of their graduate outcome by using the information about social network among students. Such a quantitative tool is a sociogram which depicts how students interact with others. The graph can be easily generated when the pattern of the connectivity among individuals is known. We apply sociogram to portray the network of a class of students in Department of Statistics – Bogor Agricultural University which represent the way they interact when they want to discuss the academic related problems. We found some interesting results are practically valuable for the one who is responsible to the study result of the students. Some results are not new, but this approach could provide more informative features than conventional tables or such things.Keywords : sociogram, social network analysi
PEMODELAN KASUS DEMAM BERDARAH DENGUE DI JAWA TIMUR DENGAN MODEL POISSON DAN BINOMIAL NEGATIF (Dengue Fever Case Modelling in East Java with Poisson and Negative Binomial Models)
No full textThe total number of dengue fever victims in East Java can be assumed to have a Poisson distribution. The Poisson regression method can be used to model the relationship of the environmental factors and dengue fevers incidents. The model of this method assumes equidispersion, that is the equality of mean and variance of the response variables. If variance of the response variable exceeds the mean, it is called overdispersion. Negative binomial regression model is used to overcome the overdispersion. Negative binomial regression model shows that the quantity of dengue fever victims in every kabupaten (district) is influenced by the quantity of flood and the quantity of malnutrition victims. Negative binomial regression shows that the increasing number of flood will enhance the quantity of dengue fever victims in East Java district whereas the increasing quantity of malnutrition victims will enhance the quantity of dengue fever victims in East Java district. Keywords : Poisson regression, negative binomial regression, overdispersio
Poling Dan Penerapannya Di Indonesia
No full textPoling yang akhir-akhir ini marak dan seringkali diselenggarakan oleh berbagai media elektronik, media cetak, dan lembaga-lembaga riset untuk melihat pendapat masyarakat terhadap suatu permasalahan yang sedang dihadapi. Hasil poling dari satu penyelenggara dengan penyelenggara lain seringkali berbeda karena metode yang digunakanpun berbeda. Poling yang diselenggarakan televisi biasanya menggunakan metode penarikan contoh tidak berpeluang (unscientific polling), sedangkan poling yang diselenggarakan oleh lembaga riset biasanya menggunakan metode penarikan contoh berpeluang (scientific polling). Sumber bias dari kebanyakan poling yang diselenggarakan disebabkan oleh metode penarikan contoh yang kurang tepat, urutan dan kualitas pertanyaan yang kurang baik, serta tingkat pengetahuan dan kepedulian responden yang rendah. Kata kunci : Poling, unscientific polling, scientific pollin
There have been two main topics developed by statisticians in a survey, i.e. sampling techniques and estimation methods. The current issues in estimation methods related to estimation of a particular domain having small size of samples or, in more extreme cases, there is no sample available for direct estimation. Sample survey data provide effective reliable estimators of totals and means for large area and domains. But it is recognized that the usual direct survey estimator performing statistics for a small area, have unacceptably large standard errors, due to the circumstance of small sample size in the area. The most commonly used models for this case, usually in small area estimation, are based on generalized linear mixed models. Some time happened that some surveys are carried out periodically so that the estimation could be improved by incorporating both the area and time random effects. In this paper we propose a state space model which accounts for the two random effects and is based on two equation, namely transition equation and measurement equation. Based on a evaluation criterion, the proposed hierarchical Bayes estimator turns out to be superior to both estimated best linear unbiased prediction (BLUP) and the direct survey estimator. The posterior variances which measure accuracy of the hierarchical Bayes estimates are always smaller than the corresponding variances of the BLUP and the direct survey estimates.
No full textThere have been two main topics developed by statisticians in a survey, i.e. sampling techniques and estimation methods. The current issues in estimation methods related to estimation of a particular domain having small size of samples or, in more extreme cases, there is no sample available for direct estimation. Sample survey data provide effective reliable estimators of totals and means for large area and domains. But it is recognized that the usual direct survey estimator performing statistics for a small area, have unacceptably large standard errors, due to the circumstance of small sample size in the area. The most commonly used models for this case, usually in small area estimation, are based on generalized linear mixed models. Some time happened that some surveys are carried out periodically so that the estimation could be improved by incorporating both the area and time random effects. In this paper we propose a state space model which accounts for the two random effects and is based on two equation, namely transition equation and measurement equation. Based on a evaluation criterion, the proposed hierarchical Bayes estimator turns out to be superior to both estimated best linear unbiased prediction (BLUP) and the direct survey estimator. The posterior variances which measure accuracy of the hierarchical Bayes estimates are always smaller than the corresponding variances of the BLUP and the direct survey estimates
TINJAUAN EMPIRIK TERHADAP DUGAAN GALAT BAKU NILAI TENGAH YANG DIHASILKAN PROC SURVEYMEANS
Full text linkPenarikan contoh yang semakin kompleks berimplikasi pada proses perhitungan galat baku dugaan parameter semakin rumit. Kesulitan menentukan galat baku ini sering menyebabkan para analis dan peneliti menggunakan formula yang didasarkan pada teknik penarikan contoh acak sederhana. SAS menyediakan PROC SURVEYMEANS yang menghasilkan galat baku dengan formula yang disesuaikan dengan penarikan contohnya. Penelitian ini menunjukkan secara empiris bahwa galat baku dugaan parameter yang dihasilkan oleh PROC SURVEYMEANS memiliki tingkat akurasi yang baik. Indikasi ini ditunjukkan oleh selang kepercayaan yang tidak memuat nilai parameter sebenarnya mendekati tingkat kesalahan (a) yang digunakan
ANALISIS KONJOIN: METODE FULL PROFILE DAN CBC UNTUK MENELAAH PERSEPSI MAHASISWA TERHADAP PILIHAN PEKERJAAN
Full text linkTulisan ini membahas perbandingan analisis konjoin metode full profile dan metode CBC. Metode full profile merupakan metode yang klasik dan cukup mudah diterapkan terutama dalam pembuatan disain pengumpulan dan analisis data, tetapi cukup merepotkan dalam tahap pengumpulan data. Sedangkan Metode CBC, walaupun agak sulit dalam disain pengumpulan dan analisis datanya tetapi pada saat pengumpulan datanya relatif lebih mudah dan dipandang lebih alamiah. Penerapan kedua metode ini dalam menelaah faktor yang paling dipertimbangkan oleh mahasiswa dalam memilih pekerjaan memberikan hasil yang relatif sama. Faktor utama yang berpengaruh terhadap pilihan pekerjaan mahasiswa adalah besarnya gaji pertama dan kesesuaian bidang pekerjaan dengan latar belakang pendidikannya
GENERALIZED VARIANCE FUNCTIONS FOR BINOMIAL VARIABLES IN STRATIFIED TWO-STAGE SAMPLING
Full text link 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\u27s will give poor results or even no GVF will be appropriate. Keywords : Generalized Variance Functions, Stratified Two-Stage Samplin
PENERAPAN PEMBOBOTAN KOMPONEN UTAMA UNTUK PEREDUKSIAN PEUBAH PADA ADDITIVE MAIN EFFECT AND MULTIPLICATIVE INTERACTION (Application of Weighted Principal Component for Variable Reduction in Additive Main Effect and Multiplicative Interaction)
Full text linkIndonesia is the country with the largest level of rice consumption in the world. Therefore, it need to be done an effort to increase the production of rice. One way to increase rice production is land management as well as conducting an intensive new superior varieties which has a high yield. Hybrid rice is a type of rice which has a higher result among superior varieties. Hybrid rice breeding can be done with multi-locations trials that involves two main factors, plant and environmental conditions. AMMI (Additive Main Effects and Multiplicative Interaction) is a method of multivariate used in plant breeding research to examine the interaction of genotype × environment on multi-locations trials. Generally, AMMI analysis is still using a single response. Whereas, the adaptation level of the plant is not only seen from the aspect of its yield. Therefore, this study based on combined response using AMMI analysis. The Data in this study is secondary data multi-locations trials on hybrid rice planting season 2008/2009 which involved four sites and 12 genotype. The measured response are = yield (ton/ha), = 1000 grain weight (gram), = the number of penicles per m2, dan = length of penicle (cm). The merger of response using weighted method by principal component. AMMI analysis with as response produce five stable genotypes in any location, that are IH804, IH805, IH806, Hibrindo, and Ciherang. AMMI is also generating specific genotypes are those that perform good adaptability at certain environment condition. IH802, IH803, and IH809 genotypes in Jember planting season 2, IH808 and Maro genotypes in Ngawi. Keywords : AMMI, the merger of response, weighted principal component metho
Analyzing The Consumer’s Rice Price using Multiple Linear Regression and X-12 ARIMA
No full textRice is one of the main foods in Indonesia. A change of rice price will cause a major effect in the lives of consumers. On the other hand, there are so many factors that influence the rice price. Thus finding key factors which are significant to the rice price, as well as forecasting the consumer’s rice price are needed in order to maintain the stabilization of rice price. The second objective is to find key factors which influence the rice price by using multiple linear regression models. The parameters were estimated by ordinary least square methods. There are 6 variables that are significant at α=5%, which are the consumer’s rice price at the previous period, rice production at the current and previous period, farmer’s GKP price, realization of domestic stock, and total rice import. The rice price will increase if the GKP price and realization of domestic stock increase whereas total rice import and the consumer’s rice price at the previous period have negative influences towards the rice price. The impact of imported rice is negative towards domestic rice. This condition will also drive negative effect towards the farmer’s income, in this case the price does not meet the farmers cost for production. To protect the farmers, the government applied a 430.00 Rp/Kg imported rice fee but this is not effective to decrease the amount of imported rice. In this model rice production at the current and previous period have positive signs, contradictory to the microeconomic theory where when the rice production increases, there will be an excess supply and the price will drop. That condition will occur only if the commodity is a free commodity and the rice is at the sufficiency level but in Indonesia, rice is affected by the government’s policy and the rice productivity is left behind by the demand. Forecasting the consumer’s rice price for the next five years was the last objective of this research. ARIMA Box–Jenkins Method, X-12 ARIMA, Winter’s Method, and Trend Analysis were compared to find the best statistical model to forecast the consumer’s rice price. X-12 ARIMA turns out to be the best method because it has the smallest MAPE, MAD, and MSD value. This result is a satisfactory because according to Findley et al. (1998) X-12 ARIMA has the capability to adjust seasonal and trading day factors which usually causes fluctuations in an economic time series data. Keyword : X-12 ARIMA