Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON DATA SCIENCE AND OFFICIAL STATISTICS

Estimation of Education Indicators in East Java Using Multivariate Fay-Herriot Model Novia Permatasari; Azka Ubaidillah
Proceedings of The International Conference on Data Science and Official Statistics Vol. 2021 No. 1 (2021): Proceedings of 2021 International Conference on Data Science and Official St
Publisher : Politeknik Statistika STIS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34123/icdsos.v2021i1.51

Abstract

Education is an important aspect in improving human resources. Data availability of education indicators in a low administrative level is needed as a basis for education planning in that region. The problem of sample size when provide a low administrative level data can be overcome by indirect estimation, namely Small Area Estimation (SAE). SAE is able to increase the effectiveness of the survey sample size by using the strength of neighbouring areas and information from auxiliary variables related to the variables of interest. We obtain simulation study to compare multivariate model to univariate model and implement multivariate model to estimate three education indicators which are obtained from the National Socio-Economic Surveys by Statistics Indonesia. Simulation results are in line with previous studies, where the multivariate Fay-Herriot model with p variable has smaller of mean squares error (MSE) than the univariate model. The model implementation to estimate CrudeParticipation Rate (APK), School Participation Rate (APS), and Pure Participation Rate (APM) also shows that the multivariate model produces smaller RRMSE than the direct estimates. It can be concluded that multivariate model is able to produce more efficient estimates than direct estimation and univariate model.
R Package Development for Difference Benchmarking in Small Area Estimation Fay-Herriot Model Zaza Yuda Perwira; Azka Ubaidillah
Proceedings of The International Conference on Data Science and Official Statistics Vol. 2021 No. 1 (2021): Proceedings of 2021 International Conference on Data Science and Official St
Publisher : Politeknik Statistika STIS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34123/icdsos.v2021i1.69

Abstract

In recent decades, the use of small area estimation (SAE) for producing official statistics has been widely recognized by many National Statistics Offices including BPS-Statistics Indonesia. For official statistics usage, the aggregation of small area estimates is expected to be numerically consistent and more efficient than the aggregation of the unbiased direct estimates that cannot be guaranteed by Fay-Herriot model. Simulation experiments are performed to assess the behaviour of the difference benchmarking method Fay-Herriot model and to compare the mean squared error (MSE). The result shows that the difference benchmarking method can produce a consistent aggregation towards the direct estimation. Furthermore, an R package was built to implement the method that is easier to be used and is already available in the CRAN website. The package has been evaluated using validity (simulation), performance, case studies, and usability tests. These evaluations show that the package is suitable for use. Implementation of the methodology is also be applied to estimate average household consumption per capita expenditure in districts in D.I. Yogyakarta province, Indonesia 2019
Big Data for Small Area Estimation: Happiness Index with Twitter Data Sheerin Dahwan Aziz; Azka Ubaidillah
Proceedings of The International Conference on Data Science and Official Statistics Vol. 2021 No. 1 (2021): Proceedings of 2021 International Conference on Data Science and Official St
Publisher : Politeknik Statistika STIS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34123/icdsos.v2021i1.248

Abstract

Data availability for small area level is one of the keys to the success of regional development. However, direct estimation of small areas can produce high error due to inadequate sample sizes so the estimation is not reliable. One of alternative solution to this problem is to use the Small Area Estimation (SAE) method which can improve precision by "borrows strength" of the corresponding region information or auxiliary variable information that is strongly related to the response variable. This study uses two SAE models, namely SAE EBLUP Fay-Herriot model with auxiliary variables Podes data and SAE with Error Measurement with auxiliary variable Twitter data. Estimation results using the SAE method are better than direct estimates. This is shown by the RSE value which produced from SAE method, both the EBLUP model and Measurement Error, is smaller than the direct estimate. Therefore, big data can be used as an alternative variable in the SAE model because the data is available in real-time, covers up to the smallest area, and relatively low cost.