TY - JOUR
TI - PENERAPAN REGRESI LINIER MULTIVARIAT PADA DISTRIBUSI UJIAN NASIONAL 2014 (Pada Studi Kasus Nilai Ujian Nasional 2014 SMP Negeri 1 Sayung)
AU - Nurani, Vica; Sudarno, Sudarno; Rahmawati, Rita
IS - Vol 4, No 3 (2015): Wisuda periode Agustus 2015
PB - Jurusan Statistika UNDIP
JO - Jurnal Gaussian
PY - 2015
SP - 697
EP - 704
UR - http://ejournal-s1.undip.ac.id/index.php/gaussian/article/view/9550
AB - National Exam is a measurement and assessment activities accession of national competency standards on specific subjects as well as a requirement that a student continue to pursue higher education. If we want to know the relationship between national exam score and semester score using multivariate linear regression analysis. Multivariate linear regression is the linear regression model with more than one response variables Y correlated and one or more predictor variables X. In the multivariate linear regression analysis, model selection is the important thing. This is because the selection of the best models in the multivariate linear regression analysis depends on the number of predictor variables involved in the model. The purpose of this study was to determine the best model in the multivariate linear regression analysis using the criteria of Mean Square Error (MSE). The result showed using MSE criterion obtained the best model with the smallest MSE value for 17424540. The best model obtained consists of six predictor variables and four response variables. The effect from response to predictor is 74,512%.Â Keywords : National Exam, Multivariate Linear Regression, MSE Criterion, Best Model.National Exam is a measurement and assessment activities accession of national competency standards on specific subjects as well as a requirement that a student continue to pursue higher education. If we want to know the relationship between national exam score and semester score using multivariate linear regression analysis. Multivariate linear regression is the linear regression model with more than one response variables Y correlated and one or more predictor variables X. In the multivariate linear regression analysis, model selection is the important thing. This is because the selection of the best models in the multivariate linear regression analysis depends on the number of predictor variables involved in the model. The purpose of this study was to determine the best model in the multivariate linear regression analysis using the criteria of Mean Square Error (MSE). The result showed using MSE criterion obtained the best model with the smallest MSE value for 17424540. The best model obtained consists of six predictor variables and four response variables. The effect from response to predictor is 74,512%.Â Keywords : National Exam, Multivariate Linear Regression, MSE Criterion, Best Model.