Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Xplore: Journal of Statistics

Implementasi Metode CHAID (Chi-Squared Automatic Interaction Detection) pada Segmentasi Trend Penjualan Minuman Ringan di Indonesia Via Sulviana; Aji Hamim Wigena; . Indahwati
Xplore: Journal of Statistics Vol. 2 No. 2 (2018): 31 Agustus 2018
Publisher : Department of Statistics, IPB

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (325.657 KB) | DOI: 10.29244/xplore.v2i2.91

Abstract

Currently some outlet sells their products by looking at sales trends over a period of time to continue developing their business and devising effective marketing strategies. CHAID (Chi-Squared Automatic Interaction Detection) method is one of the efficient non-parametric statistical methods to classify any aspects that can increase the sales of soft drinks. CHAID selects significant variables based on the Chi-Square test between categories of explanatory variables with response categories. The CHAID method is used if the response variable is nominal or ordinal. This research aims to classify characteristics that characterize diversity and determine the target market that is able to maximize profits on the sales trend of various types of soft drinks by using CHAID method. Results from CHAID are tree diagrams that divide categories of response variables by segments from explanatory variables packaged into more easily understood information. CHAID method produces 11 of 20 segments that affect the trend of soft drink sales spread across big cities of Indonesia. There are 4 independent variable from segment that form, there are city, type of outlet, source of buying and payment method which accuracy that form from segmentation are 71.4%.
Aplikasi Structural Equation Modeling-Partial Least Squares dalam Menentukan Faktor yang Mempengaruhi Kinerja Karyawan Amanda Permata Dewi; I Made Sumertajaya; Aji Hamim Wigena
Xplore: Journal of Statistics Vol. 7 No. 3 (2018): 31 Desember 2018
Publisher : Department of Statistics, IPB

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

Abstract

Structural Equation Modeling (SEM combines factor and path analysis, so researchers can see the relationship between latent variables and their indicators and the relationship between latent variables. Partial Least Square is a soft modeling approach on SEM that has no assumption of data distribution and minimum number of observations which is often called SEM-PLS. The data used in this study is the performance of 70 constructions company employees. The number of observations is too small and couldn’t fulfill the data normality assumption so the analysis method used is SEM-PLS. This study applies SEM-PLS to identify the factors that influence the performance based on competence data from each of the existing employees. The results of this study indicate that both variables have a significant influence on the performance variables. The model tested in the research is good enough to explain the diversity of the performance variables with the evaluation value of Q2 of 75.24%.
Pemodelan Produksi Ayam Ras di Indonesia Menggunakan Regresi dengan Sisaan Deret Waktu Akhbamah Primadaniyah Febrin; Itasia Dina Sulvianti; Aji Hamim Wigena
Xplore: Journal of Statistics Vol. 8 No. 1 (2019): 30 April 2019
Publisher : Department of Statistics, IPB

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29244/xplore.v8i1.192

Abstract

The production of broiler chicken has fluctuated in recent years and many factors alleged to influence the production. The purpose of this study is modeling a structural equation of forecasting the production of broiler chicken. The study use a dependent variable (Y) that is production of broiler chickens (kilo ton) and five independent variables (X) consist of broiler chicken population (million), national chicken consumption (ton/year), retail price (Rp/kg), real price of corn (Rp), and real price of Kampung chicken (Rp). The variables are time series data with errors does not spread out randomly. Modeling method used and suitable to the conditions is regression with time series errors combined with ARIMA (Autoregressive Integrated Moving Average). The results of the regression analysis showed that only population variable and retail price variable are influencing the production of broiler chicken in Indonesia. Those two independent variables then modeled by a dependent variable using regression with time series errors. The best modeling is regression with time series errors ARIMA(1,1,0) with MAPE (Mean Average Percentage Error) value of 2.4%, RMSE (Root Mean Square Error) value of 39.800, and correlation value 0.980. The results has proved that the production of broiler chicken in Indonesia is influenced by those two variables.