Rough-Regression Model for Investigating Product Attributes and Purchase Decision

Regression models have been widely applied to investigate the causal relationship between independent and dependent attributes with statistical assumptions. On the other hands, not easy to achieve all statistical assumptions using these models, especially for certain areas. This paper presents rough regression model to handle the categorical data types with minimal assumptions. The proposed idea is address to solve the unclassified elements and decisive criteria in data sets. Moreover, the product attributes of Kompas newspaper are investigated and selected using rough-regression model. The result showed that three conditional attributes, namely, price, promotion, and location have positive effect to purchasing decision attribute. Proposed decisive criteria also may help decision makers or marketing management in providing information and planning precisely.