Browsing by Author "Efendi, Rustam"
Now showing 1 - 20 of 25
Results Per Page
Sort Options
Item ANALISIS FAKTOR-FAKTOR YANG MEMPENGARUHI KEMISKINAN DI PROVINSI RIAU DENGAN METODE SEM-PLS(Elfitra, 2023-04) Rahmad, Abdul; Efendi, RustamThe increase in the number of poor people occurred in March 2021 which was caused by several factors including human resources, health, and the economy. Based on this, research is needed to find out what are the factors that cause higher poverty. This study will analyze the factors that influence poverty in Riau Province using the Structural Equation Modeling-Partial Least Square (SEM-PLS) with the bootstrap method. The variables used are three exogenous independent variables consisting of health, economy and human resources and one exogenous dependent variable namely poverty. The results of the study explain that human resources and the number of cases of disease do not have a significant and positive effect on poverty. This research shows that the higher the human resources and the number of disease cases in Riau Province in 2021, the higher the poverty rate. Meanwhile, the economy does not have a significant effect on poverty, meaning that the higher the economic level in Riau Province in 2021, the lower the poverty rate in Riau Province in 2022. Human resources, health, and the economy have no effect on poverty in Riau Province in 2021.Item ANALISIS PERBANDINGAN METODE FORWARD DAN BACKWARD DALAM PEMODELAN REGRESI LINEAR BERGANDA(Elfitra, 2023-10) Tahir, Yofi Kurnia; Efendi, RustamThe issue of health insurance costs has become a major concern in the healthcare system in the USA. Determining the accurate health insurance cost is crucial for individuals and families who rely on health insurance for financing their healthcare. This research aims to compare the most effective methods for determining health insurance costs using variable selection methods in multiple linear regression models. The variable selection methods used are forward selection and backward elimination. Based on the application in the regression model, the forward selection method yielded dan RMSE values of 79.68% and 2024.295, respectively. The backward elimination method produced dan RMSE values of 96.77% and 807.175, respectively. The most effective variable selection method for determining health insurance costs in this case is backward elimination.Item DOUBLE EXPONENTIAL SMOOTHING UNTUK PERAMALAN RATARATA HARGA BERAS INDONESIA TAHUN 2022(Elfitra, 2022-10) Irvan, M.; Efendi, RustamThis study discusses the forecasting of the average price of rice at the level of trade in Indonesia, where this study uses time series analysis. The method used in this research is Holt's Double Exponential Smoothing. The results of calculations using the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) accuracy test equipment, determined that the Double Exponential Smoothing Holt method has a relatively small forecasting error with an RMSE value of 118.963 and a MAPE value of 0.86% with MAPE value criteria < 10%. This shows that the accuracy of MAPE is in the very good category.Item ESTIMATOR POLINOMIAL LOKAL UNTUK PEMODELAN HARGA EMAS DI INDONESIA MENGGUNAKAN REGRESI NONPARAMETRIK(Elfitra, 2022-05) Dani, Rudy Rama; Efendi, RustamThis paper discusses about the nonparametric regression model based on local polynomial estimators on gold prices in Indonesia. Local polynomial estimators can be obtained by minimizing Weighted Least Square (WLS). Optimal bandwidth can be obtained using the Generalized Cross Validation (GCV) method. Furthermore this methods are applied to nonparametric regression models based on local polynomial estimators in the case of gold prices in Indonesia. Gold price modeling in Indonesia using local polynomial regression with Gaussian kernel function obtained optimal bandwidth of 10. The model produces a minimum GCV score of 854882843 with the optimum order of the local polynomial is order 4. The goodness of fit model used is the Mean Absolute Percentage Error (MAPE) with MAPE 6.182%, that is smaller them 10%. Then local polynomial regression model order 4 for gold price is very accurate.Item KOMBINASI PENAKSIR RASIO DAN PENAKSIR REGRESI UNTUK RATA-RATA POPULASI MENGGUNAKAN MEDIAN DAN KOEFISIEN SKEWNESS PADA SAMPLING ACAK SEDERHANA(2014-03-25) Sakinah; Efendi, Rustam; Gamal, M. D. HIn this artikel the combination of ratio estimators and ratio regression in simple random sampling using median and coefficient of skewness are discussed, which is a review and an expansion from the article J. Subramani & G. Kumarapandiyan [American Journal of Matematics and Statistics, 2(5): 101-107]. The estimators are biased estimators and the mean square errors are determined. Estimator with the smallest mean square error is the most efficient estimator. A numerical example is given at the end of discusssion.Item MEAN SQUARE ERROR TERKECIL DARI KOMBINASI PENAKSIR RASIO-PRODUK UNTUK RATA-RATA POPULASI PADA SAMPLING ACAK BERSTRATA(2014-03-25) Kurniati, Rini; Sugiarto, Sigit; Efendi, RustamThis paper discussed about three the ratio-product estimators for mean population in the stratified random sampling. It is combinations of ratio estimator and product estimator in stratified random sampling. This paper is a review from the paper of Tailor et.al. [Communications of the Korean Statistical Society 18:111-118]. The estimators discussed are the combination of ratio-product estimator, combination of ratio-product estimator using coefficient of variation and combination of ratio-product estimator using coefficient of kurtosis. All of estimators are bias estimator. Then the mean square error (MSE) of each estimator is evaluated. Furthermore, the MSE of each estimator is compared. This comparison shows that the combination of ratio-product estimator is the most efficient, that is with the smallest MSEItem METODE REGRESI TRANSFORMASI LOGARITMIK NATURAL PADA VARIABEL NILAI EKSPOR DI PROVINSI RIAU(2021-03) Fachri, Irfan; Efendi, RustamRiau's export is dominated by non-oil and gas, especially in agriculture. Export in agriculture is a commodity of crude palm oil and its derivatives. Natural rubber is also one of the export commodities in Riau Province. The factors tested in this study are plantation sectors that become export commodities such as the amount of palm oil production, the amount of dry rubber production, and other factors namely foreign exchange rates. The method used is the logarithmic transformation regression method. In the evaluation of the forecasting results used mape value (Mean Absolute Percentage Error). The MAPE value in the multiple regression method is 46.91366 %, and the MAPE value on the regression method is multiple with a logarithmic transformation of the response variable of 52.69479%. The smaller the MAPE value, the smaller the forecast error rate. At the level of forecasting error that multiple regression is better than multiple regression with logarithmic transformation of the response variableItem METODE RIDGE REGRESSION DALAM MENGATASI DAMPAK MULTIKOLINEARITAS PADA PRODUK DOMESTIK REGIONAL BRUTO PROVINSI RIAU(2021-04) Kurniawan, Angga; Efendi, RustamMultiple linear regressions are analytical methods used to determine the relationship between a variable and another. One of the assumptions in multiple linear regression analysis is the absence of multicolinearity in the regression model. If there is a multicolinearity problem, there are several ways that can be used by using the ridge regression method which is a modification of the smallest squared method by adding the a bias constanta c. This study aims to suspect ridge regression parameters using Hoerl, Kennard & Baldwin and Lawless &Wang methods to address multicolinearity problems in the analysis of economic factors in Riau Province. The factors tested in this study are the oil and gas sector into gross regional domestic product commodities, other factors namely exports, consumer price index and rupiah exchange rate. In practice done by comparing two methods of bias rated values Hoerl, Kennard & Baldwin and metohods Lawless & Wang. . The study showed that there is at least one ridge regression method that has MSE, smaller and and value, adjusted larger than Hoerl Kennard & Baldwin method, is obtained Lawless & Wang ridge regression method with bias = 0,288.Item MODEL COX PROPORTIONAL HAZARD DENGAN PENDEKATAN EFRON DAN EXACT PADA KECELAKAAN LALU LINTAS DI PEKANBARU(Elfitra, 2022-04) Suherman, Wahyu; Efendi, RustamThis study aims to apply the Cox proportional hazard regression model to the ties event using two approach methods, ie the Efron approach and the Exact approach and find out its application in the case of traffic accidents at Pekanbaru city. The accident data was taken from the Satlantas Polrestabes Pekanbaru on 2020. There are four variables that are considered in this study, ie age, gender, helmet use, and driver’s license ownership. After testing the best Cox model, the two approaches used gave the same results where there were three significant variables, driver’s license ownership, gender, and age. Then, the proportional hazard assumption was tested, where all variables met the proportional hazard assumptionItem MODEL FUNGSI TRANSFER DENGAN DERET GANGGUAN UNTUK MERAMALKAN CURAH HUJAN KOTA PEKANBARU TAHUN 2021(Elfitra, 2022-05) Sihombing, Fenita Wulandari; Efendi, RustamRainfall in Pekanbaru has a high intensity compared to other cities in Riau Province but during 2010-2020 the intensity becomes uncertain so it is important to forecast rainfall so that planning in all areas of human life such as agriculture, development, and so on can be maximized. Rainfall forecasting using the ARIMA model is considered less significant, so it developed into an ARIMAX model with the transfer function model method, involving exogenous variables which are assumed to affect endogenous variables. This study aims to predict the rainfall in Pekanbaru using a transfer function model with a disturbance series with wind speed as the exogenous variable. The data is divided into two, namely in sample data (rainfall data and wind speed for the January-December 2020) and out sample data (rainfall and wind speed data for the January-December 2021). Based on the results of the study, it was found that the rainfall forecasting for Pekanbaru from January to December 2021 was known that the rainfall that occurred ranged from -128 mm to 561 mm in the lowhigh rainfall category with a MAPE of 14% which is a good result.Item MODEL REGRESI NONPARAMETRIK KERNEL MENGGUNAKAN ESTIMASI NADARAYA-WATSON UNTUK DATA HARGA INDEKS SAHAM GABUNGAN DI INDONESIA(Elfitra, 2022-05) Kesumah, Dwika Irmayusmita Sandra; Efendi, RustamThe composite index is one of the stock price indexes in Indonesia. In this study the kernel nonparametric regression model with the Nadaraya-Watson estimator for the composite index data. The kernel regression method is one of the methods in nonparametric regression used to estimate conditional expectations using kernel functions. The kernel function used in this study is the Gaussian kernel function. The data used is the composite stock price index in Indonesia in 2018-2019. The first step is to determine in advance the optimal bandwidth with the Generalized Cross Validation (GCV) method. Kernel regression using gaussian kernel functions obtained a bandwidth value of 82,03 with an optimal GCV of 269,42. Based on the results of the analysis to measure the goodness of the model using Mean Absolute Square Error (MAPE) of 2,71, which means that the MAPE value is in the first category which is very good.Item PEMODELAN ANGKA PARTISIPASI SEKOLAH DI PROVINSI RIAU DENGAN PENDEKATAN MULTIVARIATE ADAPTIVE REGRESSION SPLINES (MARS(perpustakaan UR, 2021-10) Nuraini, Siti; Efendi, RustamSchool Participation Rate (SPR) is a basic indicator used to measure access to education, especially for the school-age population. Riau has fluctuating SPR data for senior high school level. In this study, the nonparametric multivariate adaptive regression splines (MARS) method was used to model the high school participation rate in the Regency/City of Riau Province in 2015-2019. Factors that are thought to affect the SPR in Riau Province are human development index, gross regional domestic product, poverty level and number of schools. This study shows that the best MARS model with a minimum GCV value of 19.83156 and an R-square of 0.745368. The most influential variable for SPR in Riau Province in 2015-2019 is the poverty rate with a 100 percent interest rate.Item PEMODELAN JUMLAH KASUS KEMATIAN BAYI DI KABUPATEN KUANTAN SINGINGI MENGGUNAKAN REGRESI BINOMIAL NEGATIF(Elfitra, 2023-04) Sari, Luci Manda; Efendi, RustamInfant mortality is one of the commonly used mortality indicators in Indonesia. Cases of infant mortality reflect the level of social welfare and health status of a nation for the realization of health development in a country. One of the ways to reveal the high number of infant mortality cases is by knowing the most significant factors that cause infant mortality. The results obtained from this study experienced overdispersion in Poisson regression, so that in the analysis using negative binomial regression, a significant variable was obtained, namely the number of medical personnel with the best AIC in negative binomial regression for a significant variable in reduction II with an AIC value of 60.42Item PEMODELAN PERSENTASE PENDUDUK MISKIN DI PROVINSI RIAU DENGAN PENDEKATAN REGRESI NONPARAMETRIK DERET FOURIER(Elfitra, 2022-12) Hamdani, Khairul; Efendi, RustamPercentage of poor people is an indicator used to measure poverty in an area. Riau Province has data on the percentage of poor people that fluctuates. In this study, the Fourier series nonparametric regression method was used to model the percentage of poor people in the Regency/City Riau Province in 2014-2018. Factors that are thought to influence the percentage of poor people in Riau Province are the open unemployment rate, per capita expenditure, and the average length of schooling. This study resulted in the best Fourier series model at 𝐾 = 4 with a minimum GCV value of 33.50 with 16 model parameters and produces a MAPE value of 17.61%, which means that the MAPE value is in the second category, which is good.Item PEMODELAN PRODUKSI PADI DI INDONESIA DENGAN PENDEKATAN MIXED GEOGRAPHICALLY WEIGHTED REGRESSION(2021-03) Muslihah, Ririn Ridania; Efendi, RustamThis paper discusses about rice production modeling in Indonesia with Mixed Geographically Weighted Regression (MGWR) approach. The MGWR model is a combination of multiple linear regression models and Geographically Weighted Regression (GWR), which model variables that are global and local. The estimated parameters of the MGWR model are obtained using the Weighted Least Square (WLS) method. The optimum bandwidth is selected by using the Cross Validation (CV) method. Rice production in Indonesia is influenced by factors such as harvest area, rainfall, fertilizer, and farmers. MGWR model produces variables of local nature is the area of harvest, while the variables that are global are rainfall, fertilizer and farmers. MGWR model is a best model used to model rice production data in Indonesia based on Akaike Information Criteria Corrected valueItem PEMODELAN REGRESI SPLINE PADA FAKTOR YANG MEMPENGARUHI TINGKAT KEMISKINAN DI INDONESIA(Elfitra, 2023-10) Zilmuarif, Firsan; Efendi, RustamIndonesia is one of the countries that is still struggling to overcome problems related to poverty. Based on official statistical news, in March 2020 the percentage of the poor population was recorded at 9.78%, which has increased periodically until September 2020 at 10.19% and in March 2021 at 10.14%, which is around 27.54 million. Although at the end of March 2022 the percentage of poor people in Indonesia decreased by 0.17% against September 2021, which is a decrease of around 0.34 million people. This study aims to model poverty in Indonesia with factors that are thought to affect it using spline nonparametric regression methods. This study obtained the best model, which is spline regression using a combination knot point (2,3,3,1,1) with a minimum GCV of 21.315. Of the five independent variables that have been tested, all variables have a significant effect on the poverty rate in Indonesia in 2021, namely the IPM, PDRB, TPT, and RLS variables.Item PENAKSIR DUAL RASIO-CUM-PRODUK UNTUK RATA-RATA POPULASI DENGAN MENGGUNAKAN KOEFISIEN KORELAS(2014-03-25) Ilham, Ronni Vel; Efendi, Rustam; Sirait, HaposanThis article discusses three ratio-cum-product estimators forthe population mean using the coefficient of correlation on simple random sampling i.e. ratio-cum-product estimator, dual ratio-cum-product estimator, and combination dual ratio-cum-product estimator. Thethree estimators are biased estimators. Then themean square errors are Determined Furthermore, these mean square errors are compared to the mean square error of each estimator.This comparison shows that the combination dual ratio-cum- product estimator using coefficient of correlationis more efficient than other estimators.Item PENAKSIR RASIO UNTUK RATA-RATA POPULASI MENGGUNAKAN SAMPLING ACAK SEDERHANA DAN SAMPLING BERPERINGKAT(2013-03-16) Suroso, Ryan Ariesta Ramli; Adnan, Arisman; Efendi, RustamRatio estimators of the population mean using simple random sampling and ranked set sampling that suggested here are two ratio estimators involving the first and the third quartiles parameters. These estimators are biased estimators, so that the Mean Square Error is calculated for each estimators to obtain the efficient estimators. Therefore, the simulation using the first and third quartiles parameters is carried out. The simulation showed that the first quartiles parameters are more efficient than the third quartileItem PENAKSIR RASIO UNTUK RATA-RATA POPULASI PADA SAMPLING ACAK BERSTRATA ADAPTIF CLUSTER(2014-03-25) Ardini, Dita; Efendi, Rustam; BustamiThis article discusses two ratio estimators for the population mean in stratified adaptive cluster random sampling, the ratio estimator for the population mean and ratio estimator for the population mean, both proposed by Nipaporn Chutiman published in Thailand Statistician Journal [1]. All estimators are biased estimators. Then the mean square errors are determined for each estimator. Futhermore, the mean square error of each estimator will be compared to obtain an efficient ratio estimator. An numerical example is given at the end of the discussion.Item PENAKSIR RASIO-PRODUK EKSPONENSIAL YANG EFISIEN UNTUK RATA-RATA POPULASI PADA SAMPLING ACAK BERSTRATA(2014-03-25) Nuryalita, Dessy; Efendi, Rustam; Sirait, HaposanThis paper discusses three estimators for the population mean in the stratified random sampling, the exponential ratio estimator and exponential product estimator proposed by Singh [4] and combination exponential ratio-product estimator proposed by Singh and Vishwakarma [3]. The three estimators are biased estimators. The most efficient estimator is an estimator that has the smallest mean square error obtained by comparing the mean square error of each estimator.