Yuliani Setia Dewi
Program Studi Magister Matematika, FMIPA, Universitas Jember

Published : 20 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 20 Documents
Search

ANALISIS REGRESI KELAS LATEN UNTUK DATA KATEGORIK DENGAN SATU KOVARIAT Haeruddin, Haeruddin; Tirta, I Made; Dewi, Yuliani Setia
BERKALA SAINSTEK Vol 1, No 1 (2013)
Publisher : My Home

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (181.268 KB)

Abstract

Analisis regresi kelas laten merupakan analisis multivariat untuk data kategorik. Estimasi parameter pada analisis regresi kelas laten menggunakan algoritma EM (ekspektasi-maksimisasi) yang dilanjutkan dengan metode Newton-Raphson. Dalam penelitian ini, analisis regresi kelas laten digunakan untuk mengklasifikasikan responden berdasarkan persepsinya terhadap peluang (opportunity) dan ancaman (treath) bagi distributor produk Unilever, PT. Panahmas Dwitama Distrindo Regional Jember. Lamanya responden berlangganan terhadap distributor ini dijadikan sebagai kovariat. Hasil analisis menunjukkan bahwa berdasarkan persepsinya terhadap opportunity, responden dikelompokkan menjadi tiga kelompok, sedangkan terhadap treath dikelompokkan menjadi dua kelompok.
POLA-POLA JALUR PADA PATH ANALISYS UNTUK ANALISIS FAKTOR-FAKTOR YANG BERPENGARUH TERHADAP NILAI UN SMA DI KABUPATEN LUMAJANG Isdarmawan, Agus; Tirta, I Made; Dewi, Yuliani Setia
KadikmA Vol 4, No 1 (2013): April 2013
Publisher : KadikmA

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (618.205 KB)

Abstract

Abstract. Path analysis is a technique to analyze the effect of free and bound variables in which every variable correlates or associates with cause and effect directly or indirectly. This study was conducted to determine some factors which influenced National Examination at Senior High School in Lumajang. The Data were analyzed using path analysis. The results of the study were explained as follows: 1. The correlation of variables in path analysis followed the pattern of direct, indirect and mixed. 2. Path analysis could be applied to the analysis of the relationship between exogenous variables (Practical Training (X1), Assignment (X2), and Daily Test (X3)) with endogenous variables (Mid-Term Test (Y1), Final-Term Test (Y2), and National Examination (Z)). Daily Test (X3) contributed directly to Mid-Term Test (Y1). On the other hand, Practical Training (X1) and Daily Test (X3) did not contribute significantly to the Final-Term Test (Y2). 3. Assignment (X2) has direct and indirect influence on National Examination (Z) through Final-Term Test (Y2). 4. Daily Test (X3) did not have a direct influence to Final-Term Test (Y2) but it had a direct impact either through National (Z or through Mid-Term Test (Y1) and Final-Term Test (Y2) which contributed 19.6% of the total site. The direct contribution of Mid-Term Test (Y1) to National Examination (Z) was the highest direct contribution in this study with 40% of the total site. While, the contribution of Practical Training (X1), Assignment (X2), Daily Test (X3), Mid-Term Test (Y1), and Final-Term Test (Y2) simultaneously influenced National Examination (Z) with 93.5% . Abaut 6.5% was influenced by the other factors which could not be described in this study. Key Words : National Examination, Path Analysis, Variable Exogenous, endogenous variables
PERAMALAN PERTUMBUHAN PENDUDUK KABUPATEN SITUBONDO DENGAN MODEL ARIMA, DERET ARITMATIK, DERET GEOMETRI DAN DERET EKSPONENSIAL “THE FORECASTING GROWTH OF THE POPULATION IN SITUBONDO BY USING ARIMA, ARITMATICS, GEOMETRICS AND EXPONENTIAL” As’ad, A; Tirta, I Made; Dewi, Yuliani Setia
KadikmA Vol 4, No 1 (2013): April 2013
Publisher : KadikmA

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (848.296 KB)

Abstract

Abstract. ARIMA models as the population forecasting in Situbondo is a model of ARIMA(3, 3, 3) and mathematically, it is stated as; =2,445–1,6632–0,148+0,9732–1,0746+ 0,4676+– 1,0635. Forecasting the population in Situbondo is 667646 people in 2012 and in 2013 is 677852 people. Some other approaches in determining the population is the Arithmetic growth formula, the result of forecasting in 2012 is 657540 people and in 2013 is 661626 people, Based on Geometric growth formula, the result of forecasting in 2012 is 19696459 people and in 2013 is 35211214 people and Based on Exponential growth formula the result of forecasting in 2012 is 657611 people and in 2013 is 661799 people. If we compare the data of the forecasted result of ARIMA model with the Aritmatics growth formula and Exponential growth formula, show that the data of the population with the last ten actual data is relatively similiar.The closed last ten actual data forecasting of population is the aritmatics growth formula, whereas the data of the population result for next two year based on the Geometric growth formula got the forecasted result which is different from the forecasted result of ARIMA model, Aritmatics growth formula and Exponential growth formula. Key Words:forecasting, arima models, arithmetic, geometric, exponential
Analisis Pengaruh Kompetensi Tenaga Guru dan Kompetensi Kepala Sekolah Terhadap Capaian Standar Nasional Pendidikan Kasmuri, Kasmuri; Tirta, I Made; Dewi, Yuliani Setia
Prosiding Seminar Matematika dan Pendidikan Matematik Vol 1, No 1 (2014): Prosiding Seminar Nasional Matematika 2014
Publisher : Prosiding Seminar Matematika dan Pendidikan Matematik

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Standar kompetensi guru ini dikembangkan secara utuh dari empat kompetensi utama (pedagogik, kepribadian, sosial, dan profesional) dan keempat kompetensi tersebut terintegrasi dalam kinerja guru. Sedangkan kepala sekolah sebagai pelaksana kepemimpinan pendidikan di sekolah harus memiliki kemampuan dan ketrampilan yang menggambarkan tugas dan peranan kepala sekolah dalam  penerapannya dituangkan dalam kompetensi kepala sekolah (kepribadian, manajerial, kewirausahaan, supervisi dan sosial). Sementara itu standar nasional pendidikan yang meliputi delapan standar (isi, proses, kelulusan, pendidik dan tenaga kependidikan, sarana dan prasarana, pengelolaan, pembeayaan dan penilaian) adalah kriteria minimal tentang sistem pendidikan di  Indonesia yang harus dicapai. Tujuan penelitian ini adalah menganalisis pengaruh kompetensi guru dan kompetensi kepala sekolah terhadap capaian standar nasional pendidikan serta mengetahui indikator-indikator yang paling dominan dalam mengukur peubah laten antara kompetensi kepala sekolah, kompetensi guru terhadap  pencapaian standar nasional pendidikan tingkat sekolah menegah di Kabupaten Banyuwangi. Data yang digunakan adalah data nilai kinerja guru, nilai kinerja kepala sekolah dan nilai pencapaian standar nasional pendidikan (akreditasi sekolah), pada sekolah menengah di Kabupaten Banyuwangi. Metode analisis yang dipakai adalah covarian based SEM dengan estimasi maximum likelihood. Hasil yang diperolah adalah pengaruh kompetensi kepala sekolah berpengaruh kuat terhadap kompetensi guru, dan kompetensi guru juga berpengaruh kuat terhadap standar nasional pendidikan. Sedangkan  indikator yang paling dominan dari varibel laten kompetensi kepala sekolah adalah kompetensi kewirausahaan, untuk variabel laten kompetensi guru adalah kompetensi kepribadian, sedangkan untuk variabel laten standar nasional pendidikan adalah standar sarana dan prasarana.
Approach Generalized Structured Component Analysis (GSCA) Method for Structural Equation Modeling Unidimensional Susanti, Nawal Ika; Tirta, I Made; Dewi, Yuliani Setia
Prosiding Seminar Matematika dan Pendidikan Matematik Vol 1, No 1 (2014): Prosiding Seminar Nasional Matematika 2014
Publisher : Prosiding Seminar Matematika dan Pendidikan Matematik

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

There are two types of Structural Equation Modeling is covarience or CB-SEM and variance or Partial Least Square SEM. The two types have advantages and disadvantages of each so Hwang & Takane propose a new method, namely the Generalized Structured Component Analysis (GSCA) which is a method that has been developed to complement the existing deficiencies in the Partial Least Square. Researchers using the GSCA for structural model factors affecting the nutritional status of children under five who are unidimensional structural equation. GSCA method in estimating the parameters using the method of Alternating Least Squares (ALS) and to estimate the standard error of the parameter estimates using the bootstrap method. The results of this study are all variables that indicator is a measure of valid and reliable to measure latent variables and also research model is a model that can be acceptable and in accordance with the existing conditions in the field.
PerbandinganAnalisisDiskriminan Linier, Diskriminan Linier RobustdanRegresiLogistikBiner Marino, Marino; Tirta, I Made; Dewi, Yuliani Setia
Prosiding Seminar Matematika dan Pendidikan Matematik Vol 1, No 1 (2014): Prosiding Seminar Nasional Matematika 2014
Publisher : Prosiding Seminar Matematika dan Pendidikan Matematik

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penelitian ini membandingkan analisis diskriminan linier, diskriminan linier robust dan regresi logistik biner untuk mengelompokan siswa SMA Negeri 1 Bangorejo ke dalam kelompok IPA/IPS. Data yang digunakanadalah data nilai raport dan psikotes siswa kelas X semester 2 tahun pelajaran 2012-2013 SMAN 1 Bangorejo Banyuwangi. Data yang digunakan merupakan data terkontaminasi outlier sebesar 6,70%. Untuk mengetahui performa terhadap keberadaan outlier, maka dilakukan simulasi secara berulang-ulang mengaplikasikan analisis diskriminan linier, diskriminan linier robust dan regresi logistik biner dengan besar sampel bervariasi yaitu n1=40, n2=80, n3=120 dan n4=120 responden dan besar outlier yang bervariasi yaitu 5%, 10%, 15% dan 20%. Dari hasil simulasi ditunjukkan bahwa regresi logisltik biner mempunyai ketepatan klasifikasi yang paling baik. Pengelompokan IPA atau IPS di SMA N. 1 Bangorejo dengan jumlah sampel keseluruhan (224 responden), dengan menggunakan analisis logistik biner mempunyai ketepatan klasifikasi sebesar 85,714%.
PEMODELAN JUMLAH KEMATIAN AKIBAT DIFTERI DI PROVINSI JAWA TIMUR DENGAN REGRESI BINOMIAL NEGATIF DAN ZERO-INFLATED POISSON Fittriyah, Nurul; Hadi, Alfian Futuhul; Dewi, Yuliani Setia
Prosiding Seminar Matematika dan Pendidikan Matematik Vol 1, No 1 (2014): Prosiding Seminar Nasional Matematika 2014
Publisher : Prosiding Seminar Matematika dan Pendidikan Matematik

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penyakit Difteri merupakan salah satu penyakit menular yang berbahaya, karena terdapat 37 kasus kematian dari 955 kasus. Bakteri Corynebacterium diphteriae menyerang saluran pernafasan atas, racun menyebar melalui darah dan dapat menyebabkan kerusakan jaringan di  seluruh tubuh terutama jantung dan saraf. Analisis regresi yang digunakan untuk variabel  tak bebas berupa data count adalah analisis regresi Poisson, namun sering kali terjadi over dispers pada regresi Poisson. Hal ini dapa diatasi dengan menggunakan regresi Binomial  Negatif, namun sering kali overdispersi pada data cacahan dapat disebabkan oleh excesszeros dan untuk mengatasinya digunakan regresi Zero-Inflated Poisson (ZIP). Keterkaitan antara prosentase cakupan desa/kelurahan UCI, jumlah kasus gizi buruk, prosentase masyarakat miskin dan hamper miskin, prosentase rumah tangga yang berperilaku hidup bersih dan sehat, serta jumlah puskesmas dengan banyaknya kematian akibat penyakit difteri dapat didekati dengan analisis statistika yang mengkaji tentang hubungan variable tak bebas dan variable bebas, yaitu analisis regresi. Langkah-langkah dalam penelitian ini adalah, pertama melakukan kajian pustaka tentang difteri. Kedua, melakukan pengujian model regresi Poisson pada data. Ketiga, mengidentifikasi overdispersi serta excesszeros. Keempat melakukan  pengujian  model regresi Binomial Negatif dan ZIP secara saturated dan full model dengan  bantuan program R. Langkah terakhir membandingkan nilai log-likelihood dari model yang didapatkan untuk mendapatkan model terbaik. Hasil penelitian ini menunjukkan bahwa model terbaik diperoleh dari model regresi ZIP dengan nilai log-likelihood sebesar-29,29.
OLS, LASSO dan PLS Pada data Mengandung Multikolinearitas Dewi, Yuliani Setia
Jurnal ILMU DASAR Vol 11 No 1 (2010)
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Jember

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (332.009 KB)

Abstract

Correlation between predictor variables (multicollinearity) become a problem in regression analysis. There are some methods to solve the problem and each method has its own complexity. This research aims to explore performance of OLS, LASSO and PLS on data that have correlation between predictor variables. OLS establishes model by minimizing sum square of residual. LASSO minimizes sum square of residual subject to sum of absolute coefficient less than a constant and PLS combine principal component analysis and multiple linear regression. By analyzing simulation and real data using R program, results of this research are that for data with serious multicollinearity (there are high correlations between predictor variables), LASSO tend to have lower bias average than PLS in prediction of response variable. OLS method has the greatest variance of MSEP, that is mostly not consistent in estimating the Mean Square Error Prediction (MSEP). MSEP that is resulted by using PLS is less than that by using LASSO. 
Structural Equation Modeling of the Factors Affecting the Nutritional Status of Children Under Five in Banyuwangi Region using Recursive (one-way) GSCA Tirta, I Made; Susanti, Nawal Ika; Dewi, Yuliani Setia
Jurnal ILMU DASAR Vol 16 No 1 (2015)
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Jember

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1517.157 KB) | DOI: 10.19184/jid.v16i1.534

Abstract

Structural Equation Modeling is one among popular multivariate analysis, especially applied in pschology and marketing. There are two main types of Structural Equation Modeling namely covariance-based or CB-SEM and variance-based or Partial Least Square (PLS)- SEM. Both types have advantages and disadvantage. To overcome its limitation, Generalized Structured Component Analysis (GSCA) was then proposed as an extension of PLS-SEM. In estimating the parameters, GSCA uses Alternating Least Squares (ALS) and in estimating the standard error of the parameter estimates it uses the bootstrap method. In this paper, GSCA is applied to study the causality model of Infant nutritional status, in relation with socio-economic status and infantcare status in Banyuwangi Region. The results show that both socio-economic and infantcare status have significant positive influence on infant nutritional status.Keywords:  Alternating least square, generalized structural component analysis,  nutritional status of infants,  structural equation modelling
The Efficiency of First (GEE1) and Second (GEE2) Order “Generalized Estimating Equations” for Longitudinal Data Hidayati, Rizka Dwi; Tirta, I Made; Dewi, Yuliani Setia
Jurnal ILMU DASAR Vol 15 No 1 (2014)
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Jember

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (799.799 KB) | DOI: 10.19184/jid.v15i1.553

Abstract

The approach of GEE focuses on a linear model for the mean of the observations in the cluster without full specification  the distribution of full-on observation. GEE is a marginal model where is not based on the full likelihood of the response, but only based on the relationship between the mean (first moment) and variance (second moment) as well as the correlation matrix. The advantage of  GEE is that the mean of  parameter are estimated consistently regardless whether  the correlation structure is specified correctly or not, as long as the mean has the correct specifications. However, the efficiency may be reduced when the working correlation structure is wrong. GEE was designed to focus on the marginal mean and correlation structure as nuisiance treat. Implementation of GEE is usually limited to the number of working correlation structure (eg AR-1, exchangeable, independent, m-dependent and unstructured). To increase the efficiency of the GEE, has introduced a variation called the Generalized Estimating Equations order 2 (GEE2). GEE2 has been introduced to overcome the problem that considers correlation GEE as nuisiance, by applying the second equation to estimate covariance parameters and  solved simultaneously with the first equation. This study used simulation data which are designed based on the the AR-1 and Exchangeable correlation structure, then estimation are done  using theAR1 and exchangeable. For GEE2,  estimation done by adding model for correlation link. The result is a link affects the efficiency of the model correlation is shown with standard error values ​​generated by GEE2 method is smaller than the GEE method.