Spatial Autoregressive Quantile Regression Modeling of the Distribution of Drug Users in the District Karo

Abstract

BNN data shows that an average of 50 people die from drugs every day, and Karo Regency is in second place for the distribution of drug abuse in North Sumatra after Medan City. Variables that have this risk include drug factors, namely availability and ease of obtaining drugs, individual factors, environmental factors, family factors, and social factors. Based on this, a model is needed to determine the development of the case. The SARQR model combines SAR modeling with quantile regression (QR). Combining the SAR model with quantile regression produces a model that is good for overcoming the problems of dependency and heterogeneity in modeling spatial data and is resistant to outlier data. This research aims to determine the Spatial Autoregressive Quantile Regression Model for the distribution of drug users in the Karo Regency. The type of research used is quantitative research. The data type used is secondary data, namely, the kind of data already existing, and the data source used in this research is drug users in the Karo district. The research results show that the Spatial Autoregressive Quantile Regression model for the distribution of drug users in Karo Regency obtained estimation results for the distribution parameters of drug users using a significance test. This model explains that the factors that significantly influence drug abuse are age, gender, occupation, and other underlying factors.