Performance Analysis and Discrimination Procedure of Two-Group Location Model with Some Continuous and High-Dimensional of Binary Variables

This research’s primary goal was to evaluate the performance analysis of recently constructed smoothed location models (SLMs) for discrimination purposes by combining two kinds multiple correspondence (MCA) handle high dimensionality problems arising from binary variables. A previous study SLM, together with MCA as well principal component (PCA), displayed that misclassification rate still very respect a large number Thus, new SLMs are in this paper solve particular problem. The first model results combination SLM Burt (denoted SLM+Burt), and second one is joint SLM+JCA). findings showed both performed all sample sizes (n) variables (b) under investigation, except n=60 b=25 SLM+JCA model. Overall, yields greater contrast SLM+Burt Moreover, concept procedures two-group classification conducted can be extended multi-class practitioners often deal many groups complexities