Claim Missing Document
Check
Articles

Found 2 Documents
Search

PENDETEKSIAN INFLUENTIAL OBSERVATION PADA MODEL REGRESI LINIER MULTIVARIAT MENGGUNAKAN JARAK COOK TERGENERALISASI (STUDI KASUS INDIKATOR PENDIDIKAN PROVINSI JAWA TENGAH TAHUN 2010) Puti Cresti Ekacitta; Diah Safitri; Triastuti Wuryandari
Jurnal Gaussian Vol 1, No 1 (2012): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (570.668 KB) | DOI: 10.14710/j.gauss.v1i1.906

Abstract

Multivariate linear regression model is regression model with one or more response variable and one or more predictor variable, with each response variable are mutually. In multivariate linear regression model sometimes often found Influential Observation. Influential Observation give most contributing in estimating regression coefficient. For detection Influential Observation on multivariate linear regression model is used Generalized Cook’s Distance. The aim of this research is to detection any or not any Influential Observation on multivariate linear regression model of education indicator in Central Java Province with response variable are Gross Participation Rate (APK), School Participation Rate (APS), and Pure Participation Number (APM) and predictor variable is percentage of population aged 10 years and over who graduated from junior high school. Result from this research  can be explained that if the percentage of population aged 10 years and over who graduated from junior high school increase one percent, it will have an impact on increasing gross participation rate the junior high school is 1.7849 % , increasing school participation rate is 1.6275 % and   increasing pure participation number is 1.3712 %. Also, from this results were obtained two observations are included Influential observation. Elimination of the two observations are included Influential observation in the multivariate linear regression model of education indicators in Central Java, affects the regression coefficients change only and does not have a major impact on the closeness of the relationship between response variables and predictor variables in the multivariate.
Spatial Autocorrelation Analysis of East Java Stunting Prevalence Cases in 2023 Trimono; Amri Muhaimin; Ekacitta, Puti Cresti; Ardiani, Ardia Eva
Journal of Advances in Information and Industrial Technology Vol. 7 No. 1 (2025): May
Publisher : LPPM Telkom University Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52435/jaiit.v7i1.689

Abstract

Stunting is one of the chronic nutritional problems occur in East Java. In 2022, the percentage of stunting in East Java reached 19.2% and decreased to 17.7% in 2023. The less significant decrease occurred due to various factors, including malnutrition, poor sanitation, and environmental influences. This study will analyze the spatial influence on the prevalence of stunting in East Java, especially in 2023. The methods used include the Morans Index and the Local Indicator of Spatial Association (LISA). Spatial correlation analysis will help in determining the pattern of regional grouping based on stunting cases. This model works by testing whether the values of a variable at a location are related to the values of the same variable at neighboring locations, with the nature of the relationship being positive (clustering) or negative (dispersion). Using stunting prevalence data in 2023, the Moran Index = 0.3233 was obtained with a Zvalue = -1.0776. This value indicates that there is positive spatial autocorrelation, but is not significant enough. Then, through the Moran Scatterplot analysis, there are indications of regional grouping in four spatial quadrants. The results of the LISA analysis show that there are five cities/regencies included in the High-High cluster (Jember, Probolinggo City, Lumajang, Malang, and Probolinggo), one area in the Low-High cluster (Situbondo), and one area in the Low-Low cluster (Gresik). These findings indicate the existence of a spatial concentration of stunting problems that can be used as a basis for developing appropriate handling strategies by the provincial government based on regions.