A synthetic data generation procedure for univariate circular data with various outliers scenarios using Python programming language

• Main Authors: Nur Syahirah, Zulkipli, Siti Zanariah, Satari, Wan Nur Syahidah, Wan Yusoff
• Format: Conference or Workshop Item
• Language: English
• Published: IOP Publishing Ltd 2021
• Subjects: Q Science (General); QA Mathematics
• Online Access: http://umpir.ump.edu.my/id/eprint/35201/1/A%20synthetic%20data%20generation%20procedure%20for%20univariate%20circular%20data%20with%20various%20outliers%20scenarios.pdf; http://umpir.ump.edu.my/id/eprint/35201/; https://doi.org/10.1088/1742-6596/1988/1/012111
• Institution: Universiti Malaysia Pahang

Synthetic data is artificial data that is created based on the statistical properties of the original data. The aim of this study is to generate a synthetic or simulated data for univariate circular data that follow von Mises (VM) distribution with various outliers scenario using Python programming language. The procedure of formulation a synthetic data generation is proposed in this study. The synthetic data is generated from various combinations of seven sample size, n and five concentration parameters, K. Moreover, a synthetic data will be generated by formulating a data generation procedure with different condition of outliers scenarios. Three outliers scenarios are proposed in this study to introduce the outliers in synthetic dataset by placing them away from inliers at a specific distance. The number of outliers planted in the dataset are fixed with three outliers. The synthetic data is randomly generated by using Python library and package which are 'numpy', 'random' and von Mises'. In conclusion, the synthetic data of univariate circular data from von Mises distribution is generated and the outliers are successfully introduced in the dataset with three outliers scenarios using Python. This study will be valuable for those who are interested to study univariate circular data with outliers and choose Python as an analysis tool.