scholarly journals PENERAPAN GSTAR-SUR PADA JUMLAH PENUMPANG PESAWAT DOMESTIK DI BANDARA INDONESIA

2020 ◽  
Vol 9 (2) ◽  
Author(s):  
Gilang Habibie, Yundari, Hendra Perdana
Keyword(s):  
Space Time ◽  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Ordinary Least Square ◽  
Unrelated Regression

Generalized space time autoregressive (GSTAR) adalah model ruang waktu yang banyak digunakan di Indonesia. Sebagian besar penelitian model GSTAR menggunakan ordinary least square (OLS) untuk mengestimasi parameter. Namun, estimasi dengan metode OLS pada model GSTAR dengan residual saling berkorelasi akan menghasilkan estimator yang tidak efisien terutama pada data musiman. Metode estimasi yang sesuai untuk residual yang saling berkorelasi adalah generalized least square (GLS), yang biasa digunakan dalam model seemingly unrelated regression (SUR). Penelitian ini bertujuan untuk menganalisis model GSTAR-SUR dan membandingkannya dengan GSTAR-OLS dengan bobot seragam dan jarak. Data yang digunakan adalah data jumlah penumpang pesawat domestik setiap bulan di Bandara Polonia/Kualanamu, Soekarno-Hatta, Juanda dan Ngurah Rai dari Januari 2006 hingga September 2019. Hasil estimasi parameter GSTAR-SUR dengan bobot seragam adalah Polonia/Kualanamu (∅10=-0,494; ∅11=0,046), Soekarno-Hatta (∅10=-0,300; ∅11=-0,828), Juanda (∅10=-0,451; ∅11=0,033) dan Ngurah Rai (∅10=-0,198; ∅11=-0,019). Sedangkan GSTAR-SUR dengan bobot jarak menghasilkan estimasi Polonia/Kualanamu (∅10=-0,492; ∅11=0,026), Soekarno-Hatta (∅10=-0,292; ∅11=-1,186), Juanda (∅10=-0,455; ∅11=0,058) dan Ngurah Rai (∅10=-0,211; ∅11=0,017). Berdasarkan nilai MAPE GSTAR-SUR lebih baik dari GSTAR-OLS dengan nilai MAPE untuk model GSTAR-OLS adalah 12,90% pada bobot seragam dan 13,43% pada bobot jarak. Model GSTAR-SUR menghasilkan nilai MAPE 6,65% untuk bobot seragam dan 7,06% untuk bobot jarak. Model terbaik adalah GSTAR-SUR bobot seragam dengan nilai MAPE 6,65%. Kata Kunci : OLS, GLS, spacetime, korelasi eror

PENERAPAN SEASONAL GENERALIZED SPACE TIME AUTOREGRESSIVE SEEMINGLY UNRELATED REGRESSION (SGSTAR SUR) PADA PERAMALAN HASIL PRODUKSI PADI

2021 ◽  
Vol 10 (2) ◽  
pp. 241-249
Author(s):  
Leni Pamularsih ◽  
Mustafid Mustafid ◽  
Abdul Hoyyi
Keyword(s):  
Rice Yield ◽  
Space Time ◽  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Training Data ◽  
Seasonal Effect ◽  
Validation Data ◽  
Central Java ◽  
Unrelated Regression ◽  
Inverse Distance

Ordinary Least Square (OLS) is general method to estimate Generalized Space Time Autoregressive (GSTAR) parameters. Parameter estimation by using OLS for GSTAR model with correlated residuals between equations will produce inefficient estimators. The method that appropriate to estimate the parameter model with correlated residuals between equations is Generalized Least Square (GLS), which is usually used in Seemingly Unrelated Regression (SUR). This research aims to build the seasonal GSTAR SUR model as model of rice yield forecasting in three locations by using the best weighting. Weights used are binary weights, inverse distance and normalization of cross correlation. Data which used in this research are the data of rice yield per quarter in three districts in Central Java, namely Banyumas, Cilacap and Kebumen. The data from the period of January 1981 to December 2014 as training data and the period of January 2015 to December 2018 as validation data. The resulting is a model that has a seasonal effect with the autoregressive order and the spasial order limited to 1 so the model formed is SGSTAR (41)-I(1)(1)3. The best model produced is the SGSTAR SUR (41)-I(1)(1)3 model with inverse distance weighting because it fulfills both assumptions, residuals white noise and residuals normally multivariate distribution. Additionally, it has the smallest MAPE value when compared the other weighting, that is 20%. This MAPE value indicates  that the accuracy rate of forecast is accurate.Keywords: Rice yield, Seasonal, GSTAR, SUR.

scholarly journals PEMODELAN INDEKS HARGA KONSUMEN DI JAWA TENGAH DENGAN METODE GENERALIZED SPACE TIME AUTOREGRESSIVE SEEMINGLY UNRELATED REGRESSION (GSTAR-SUR)

2018 ◽  
Vol 7 (4) ◽  
pp. 337-347
Author(s):  
Mega Fitria Andriyani ◽  
Abdul Hoyyi ◽  
Hasbi Yasin
Keyword(s):  
Price Index ◽  
Consumer Price Index ◽  
Estimation Method ◽  
Space Time ◽  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Residual Correlation ◽  
Central Java ◽  
Previous Time ◽  
Unrelated Regression

The Generalized Space Time Autoregressive (GSTAR) model with Seemingly Unrelated Regression (SUR) estimation method or often called GSTAR-SUR is more efficient to be used for residual correlation than Ordinary Least Square (OLS) estimation method. The SUR estimation method utilizes residual correlation information to improve the estimated efficiency resulting in a smaller standard error. The purpose of this research is to get the GSTAR-SUR model according to Consumer Price Index (CPI) data in four regencies or cities in Central Java namely Purwokerto, Surakarta, Semarang, and Tegal. Based on the assumed white noise assumption, the smallest MAPE and RMSE averages, the best model chosen in this research is the GSTAR-SUR(11)I(1) model with the heavy of normalized cross-correlation with the average MAPE value of 0.4455% and RMSE value of 0.80582. The best model obtained explains that the CPI data in Purwokerto, Semarang, and Tegal not only influenced by the previous time but also influenced by the locations. Meanwhile, the CPI data in Surakarta is only influenced by the previous time, but it is not affected by other locations. Keywords: SUR, OLS, Consumer Price Index

scholarly journals Seemingly Unrelated Regression Approach for GSTARIMA Model to Forecast Rain Fall Data in Malang Southern Region Districts

CAUCHY ◽  
2016 ◽  
Vol 4 (2) ◽  
pp. 57 ◽  
Author(s):  
Siti Choirun Nisak
Keyword(s):  
Time Series ◽  
Covariance Matrix ◽  
Ordinary Least Squares ◽  
Space Time ◽  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Unrelated Regression ◽  
Estimate Parameter ◽  
Seasonal Data ◽  
Variance Covariance Matrix

Time series forecasting models can be used to predict phenomena that occur in nature. Generalized Space Time Autoregressive (GSTAR) is one of time series model used to forecast the data consisting the elements of time and space. This model is limited to the stationary and non-seasonal data. Generalized Space Time Autoregressive Integrated Moving Average (GSTARIMA) is GSTAR development model that accommodates the non-stationary and seasonal data. Ordinary Least Squares (OLS) is method used to estimate parameter of GSTARIMA model. Estimation parameter of GSTARIMA model using OLS will not produce efficiently estimator if there is an error correlation between spaces. Ordinary Least Square (OLS) assumes the variance-covariance matrix has a constant error 𝜀𝑖𝑗~𝑁𝐼𝐷(𝟎,𝝈𝟐) but in fact, the observatory spaces are correlated so that variance-covariance matrix of the error is not constant. Therefore, Seemingly Unrelated Regression (SUR) approach is used to accommodate the weakness of the OLS. SUR assumption is 𝜀𝑖𝑗~𝑁𝐼𝐷(𝟎,𝚺) for estimating parameters GSTARIMA model. The method to estimate parameter of SUR is Generalized Least Square (GLS). Applications GSTARIMA-SUR models for rainfall data in the region Malang obtained GSTARIMA models ((1)(1,12,36),(0),(1))-SUR with determination coefficient generated with the average of 57.726%.

scholarly journals Kesenjangan Upah Antargender di Indonesia: Bukti Empiris di Sektor Manufaktur

2018 ◽  
pp. 1-21
Author(s):  
Martesa Husna Laili ◽  
Arie Damayanti
Keyword(s):  
Manufacturing Sector ◽  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Wage Discrimination ◽  
Firm Level ◽  
Least Square Estimator ◽  
Exporting Firms ◽  
Unrelated Regression ◽  
Level Data ◽  
And Gender

Theoretically, in the labor market without discrimination, wages should be paid according to productivity. Unlike other studies that use worker level data, this study will identify gender wage discrimination using firm-level data. Using Industrial Survey Data in 1996 and 2006, the gender wage ratio and gender productivity ratio were estimated simultaneously using the nonlinear seemingly unrelated regression (NLSUR) with least square estimator. We find that there is wage discrimination against women in the manufacturing sector. After disaggregating the firms by trade orientation, we show that wage discrimination against women occurs in non-exporting firms. While in exporting firms there is no wage discrimination. ========================= Secara teori, di pasar kerja yang tidak ada diskriminasi, seharusnya upah dibayar sesuai dengan produktivitas. Berbeda dengan penelitian lain yang menggunakan data level pekerja, penelitian ini akan mengidentifikasi diskriminasi upah antargender dengan menggunakan data di level perusahaan. Dengan menggunakan data Industri Besar dan Sedang tahun 1996 dan 2006, rasio upah gender dan rasio produktivitas gender diestimasi secara simultan menggunakan metode non-linear seemingly unrelated regression (NLSUR) dengan estimator least square. Penelitian ini menemukan bukti ada diskriminasi upah terhadap perempuan di sektor manufaktur. Setelah mendisagregasi perusahaan berdasarkan status ekspor, diskriminasi upah terhadap perempuan ditemukan di perusahaan non-eksportir, sedangkan di perusahaan eksportir tidak ditemukan diskriminasi upah.

Pengaruh PDRB, Tingkat Pengangguran Terbuka dan UMR Terhadap Indeks Pembangunan Manusia di Kabupaten dan Kota Provinsi Jawa Timur

Bharanomics ◽  
2021 ◽  
Vol 2 (1) ◽  
pp. 33-46
Author(s):  
Lintang Sania ◽  
Mohammad Balafif ◽  
Nurul Imamah
Keyword(s):  
Time Series ◽  
Cross Section ◽  
Fixed Effect ◽  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Unrelated Regression

Penelitian ini bertujuan untuk membuktikan apakah Pengaruh Produk Domestik Regional Bruto, Tingkat Pengangguran Terbuka Dan Upah Minimum Regional Terhadap Indeks Pembangunan Manusia Di Kabupaten Dan Kota Provinsi Jawa Timur Tahun 2014-2019. Pengambilan data menggunakan data sekunder diambil melalui website BPS Jawa Timur, yaitu data PDRB, Tingkat Pengangguran Terbuka dan Indeks Pembangunan Manusia Tahun 2014-2019, sedangkan data UMR diambil melalui Surat Keputusan Gubernur Jawa Timur Tahun 2014-2019. Metode analisis dalam penelitian ini menggunakan analisis regresi data panel yang merupakan gabungan antara data time series dan cross section yang dianalisis dengan Model Fixed Effect (FEM) dengan penimbang Feasible Generalized Least Square-Seemingly Unrelated Regression (FGLS-SUR) yang diolah dengan aplikasi EViews 9.0 diperoleh persamaan regresi IPM = -27.22579 + 3.380970*LNPDRBit + (-0.035903)*TPTit + 4.433382*LNUMRit . Hasil penelitian ini menunjukkan bahwa variabel independen produk domestik regional bruto dan upah minimum regional berpengaruh positif signifikan terhadap Indeks Pembangunan Manusia di Provinsi Jawa Timur.Sedangkan untuk variabel tingkat pengangguran terbuka berpengaruh negatif signifikan terhadap Indeks Pembangunan Manusia.

scholarly journals MODEL GENERALIZED SPACE TIME AUTOREGRESSIVE (1,1) PADA DATA CURAH HUJAN

2020 ◽  
Vol 9 (2) ◽  
Author(s):  
Ihzal Muhaini, Dadan Kusnandar, Nurfitri Imro’ah
Keyword(s):  
Moving Average ◽  
Space Time ◽  
Least Square ◽  
Autoregressive Integrated Moving Average ◽  
Ordinary Least Square

Generalized Space Time Autoregressive (GSTAR) adalah salah satu model yang digunakan untuk menganalisis data yang mempunyai ketergantungan lokasi dan waktu. Model GSTAR menghasilkan model ruang waktu yang mengadopsi tahapan-tahapan model Autoregressive Integrated Moving Average (ARIMA). Studi kasus yang digunakan pada penelitian ini adalah data curah hujan di Kota Pontianak, Kabupaten Mempawah dan Kabupaten Kubu Raya dengan periode waktu dari bulan Januari 2008 hingga bulan Desember 2012. Penelitian ini menggunakan prinsip parsimony model, sehingga model yang digunakan adalah GSTAR(1,1). Pendugaan parameter pada model GSTAR(1,1) dilakukan menggunakan metode Ordinary Least Square (OLS) dengan bobot normalisasi korelasi silang. Hasil perhitungan nilai MAPE AR(1) dan GSTAR(1,1) terlalu besar, sehingga model tidak cocok digunakan untuk peramalan pada tiga lokasi. Kata kunci: deret waktu, GSTAR, bobot normalisasi korelasi silang

scholarly journals On Seemingly Unrelated Regression and Single Equation Estimators Under Heteroscedastic Error and Non-Gaussian Responses

2020 ◽  
Vol 5 (2) ◽  
Author(s):  
Rasaq B Afolayan ◽  
Alabi W Banjoko ◽  
Mohammad K Garba ◽  
Waheed B Yahya
Keyword(s):  
Small Sample ◽  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Response Model ◽  
Sample Sizes ◽  
Contemporaneous Correlation ◽  
Heteroscedastic Error ◽  
Unrelated Regression ◽  
Error Terms ◽  
Non Gaussian

This study investigated the efficiency of Seemingly Unrelated Regression (SUR) estimator of Feasible Generalized Least Square (FGLS) compared to robust MM-BISQ, M-Huber, and Ordinary Least Squares (OLS) estimators when the variances of the error terms are non-constant and the distribution of the response variables is not Gaussian. The finite properties and relative performance of these other estimators to OLS were examined under four forms of heteroscedasticity of the error terms, levels of Contemporaneous Correlation (Cc) with gamma responses. The efficiency of four estimation techniques for the SUR model was examined using the Root Mean Square Error (RMSE) criterion to determine the best estimator(s) under different conditions at various sample sizes. The simulation results revealed that the SUR estimator (FGLS) showed superior performance in the small sample situations when the contemporaneous correlation ( ) is almost perfect ( =0.95) with the gamma response model while MM-BISQ was the best under low contemporaneous correlation. The relative efficiencies of MM-BISQ, M-Huber and FGLS estimators over the OLS are respectively 89%, 71%, and 14% in a small sample 30) and 49%, 32% and 1% in large sample sizes  under gamma response model. The study concluded that MM-BISQ and M-Huber estimators are the most efficient estimators for modeling systems of simultaneous equations with non-Gaussian responses under either homoscedastic or multiplicative heteroscedastic error terms irrespective of the sample size.Keywords—, Contemporaneous correlation, Feasible Generalized Least Square, Heteroscedasticity, Homoscedasticity, Seemingly unrelated Regression. 

scholarly journals ANALISIS KETERKAITAN A NTAR KOMODITAS PROTEIN DENGAN MENGGUNAKAN MODEL ALMOST IDEAL DEMAND SYSTEM (AIDS)

2013 ◽  
Vol 2 (3) ◽  
pp. 162
Author(s):  
Cesa Febri Desti ◽  
Dodi Devianto ◽  
Izzati Rahmi HG
Keyword(s):  
Least Square ◽  
Seemingly Unrelated Regression ◽  
Demand System ◽  
Almost Ideal Demand System ◽  
Unrelated Regression

Penelitian ini bertujuan untuk melihat keterkaitan antar harga komoditas protein dengan menggunakan model Almost Ideal Demand System (AIDS).Objek penelitian adalah mahasiswa matematika Pasca Sarjana Universitas Andalas Padang yangmengkonsumsi komoditi sumber protein hewani meliputi : daging, ayam dan telur. Pendugaan parameter menggunakan metode Generalized Least Square (GLS) melalui persamaan Seemingly Unrelated Regression (SUR). Hasil penelitian menunjukkan proporsikonsumsi pangan yang dominan adalah komoditas ayam sebesar 0.409. Nilai elastisitas harga permintaan untuk ketiga komoditi memiliki tanda negatif, ini berarti bahwaketiga komoditi merupakan kebutuhan pokok. Elastisitas pendapatan bertanda positif,mengindikasikan bahwa ketiga komoditi adalah barang normal. Pada umumnya elastistasharga silang bertanda positif, mengindikasikan bahwa antar komoditi pangan memilikihubungan saling menggantikan.

scholarly journals Identifikasi Model Generalized Space-time Autoregressive (GSTAR) untuk Nilai Inflasi di Pulau Sulawesi

2021 ◽  
Vol 18 (1) ◽  
pp. 75-83
Author(s):  
Nur'Eni ◽  
D Lusiyanti ◽  
I Gunawan
Keyword(s):  
Forecast Model ◽  
Space Time ◽  
Least Square ◽  
Ordinary Least Square

This study aims to identify a forecast model for the value of inflation at seven locations on the island of Sulawesi, namely Palu, Makassar, Gorontalo, Kendari, Manado, Mamuju and Palopo. Estimation of the parameters of the GSTAR model is carried out using the Ordinary Least Square (OLS) method with uniform location weights. The analysis results show that the GSTAR model (1,1) can be used to predict the value of inflation in Sulawesi Island.

Sign in / Sign up

Export Citation Format

Share Document