Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Media Statistika

UTILIZATION OF STUDENT’S T DISTRIBUTION TO HANDLE OUTLIERS IN TECHNICAL EFFICIENCY MEASUREMENT Zulkarnain, Rizky; Djuraidah, Anik; Sumertajaya, I Made; Indahwati, Indahwati
MEDIA STATISTIKA Vol 14, No 1 (2021): Media Statistika
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/medstat.14.1.56-67

Abstract

Stochastic frontier analysis (SFA) is the favorite method for measuring technical efficiency. SFA decomposes the error term into noise and inefficiency components. The noise component is generally assumed to have a normal distribution, while the inefficiency component is assumed to have half normal distribution. However, in the presence of outliers, the normality assumption of noise is not sufficient and can produce implausible technical efficiency scores. This paper aims to explore the use of Student’s t distribution for handling outliers in technical efficiency measurement. The model was applied in paddy rice production in East Java. Output variable was the quantity of production, while the input variables were land, seed, fertilizer, labor and capital. To link the output and inputs, Cobb-Douglas or Translog production functions was chosen using likelihood ratio test, where the parameters were estimated using maximum simulated likelihood. Furthermore, the technical efficiency scores were calculated using Jondrow method. The results showed that Student’s t distribution for noise can reduce the outliers in technical efficiency scores. Student’s t distribution revised the extremely high technical efficiency scores downward and the extremely low technical efficiency scores upward. The performance of model was improved after the outliers were handled, indicated by smaller AIC value.
PERAMALAN CURAH HUJAN EKSTRIM DI PROVINSI BANTEN DENGAN MODEL EKSTRIM SPASIAL Anik Djuraidah; Cici Suheni; Banan Nabila
MEDIA STATISTIKA Vol 12, No 1 (2019): 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 (329.719 KB) | DOI: 10.14710/medstat.12.1.50-62

Abstract

Extreme rainfall can cause negative impacts such as floods, landslides, and crop failures. Extreme rainfall modeling using spatial extreme models can provide location information of the event. Spatial extreme models combine the extreme value theory, the max-stable process, and the geostatistical correlation function of F-madogram. The estimation of the return value on the spatial extreme models is performed using the copula approach. This research used monthly rainfall data from January 1998 until December 2014 at 19 rain stations in Banten Province. The results showed that there was a high spatial dependence on extreme rainfall data in Banten Province. The forecast in range 1.5 years showed the best result compared to other ranges (1 year, 3 years, and 5 years) with MAPE 20%. The pattern of extreme rainfall forecasting was similar to its actual value with a correlation of 0.7 to 0.8. The predicted location that has the highest extreme rainfall was the Pandeglang Regency. Extreme rainfall forecasting at 19 rain stations in Banten Province using spatial extreme models produced a good forecasting.
PEMODELAN KEMISKINAN DI JAWA MENGGUNAKAN BAYESIAN SPASIAL PROBIT PENDEKATAN INTEGRATED NESTED LAPLACE APPROXIMATION (INLA) Retsi Firda Maulina; Anik Djuraidah; Anang Kurnia
MEDIA STATISTIKA Vol 12, No 2 (2019): 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 (388.748 KB) | DOI: 10.14710/medstat.12.2.140-151

Abstract

Poverty is a complex and multidimensional problem so that it becomes a development priority. Applications of poverty modeling in discrete data are still few and applications of the Bayesian paradigm are also still few. The Bayes Method is a parameter estimation method that utilizes initial information (prior) and sample information so that it can provide predictions that have a higher accuracy than the classical methods. Bayes inference using INLA approach provides faster computation than MCMC and possible uses large data sets. This study aims to model Javanese poverty using the Bayesian Spatial Probit with the INLA approach with three weighting matrices, namely K-Nearest Neighbor (KNN), Inverse Distance, and Exponential Distance. Furthermore, the result showed poverty analysis in Java based on the best model is using Bayesian SAR Probit INLA with KNN weighting matrix produced the highest level of classification accuracy, with specificity is 85.45%, sensitivity is 93.75%, and accuracy is 89.92%.