Descriptive analysis of circular data with outliers using Python programming language

  • N.S. Zulkipli Universiti Malaysia Pahang
  • S.Z. Satari Universiti Malaysia Pahang
  • W.N.S. Wan Yusoff Universiti Malaysia Pahang
Keywords: Circular data; descriptive analysis; python; programming language; outlier


Descriptive statistics are commonly used in data analysis to describe the basic features of raw data. Descriptive summaries enable us to present the data in a more simple and meaningful way so that the interpretation will be easier to understand. The descriptive analysis of circular data with outliers is discussed in this study. Circular data is different from linear data in many aspects such as statistical modeling, descriptive statistics and etc. Hence, unlike linear data, the availability of statistical software specialises in analysing circular data is very limited. Python is a programming language which frequently used by data analysts nowadays. However, the package for circular statistics is not fully developed and it is not ready to use like in Splus or R programming language. In this study, the descriptive analysis of circular data is performed using the in-demand programming language, Python. Descriptive statistics of the circular data especially with the existence of outliers are discussed and the proposed Python code is available to use.