Analysis of Malaysian Wind Direction Data Using ORIANA

Abstract

Of late, analysis of circular variables or directional data have gained much attention as they describe most of the environmental phenomena such as waves, wind gust, tornados and others. Unlike linear data, the availability of statistical software dedicated to analyze of directional data is scarce. Furthermore, the analyses are limited to descriptive summary, point estimation and comparison of means. This could partly due to the difficulty in statistical analysis of circular data because of disparate topologies between circle and straight line. For example, if the angles are recorded in the range -π, π) radian or (0, 360, then the direction close to the opposite end-points are near neighbours in a metric if we refer to the topology of circle, but maximally distant in linear metric. Thus, the distance between 350 and 15 angular degrees is more commonly thought as 25 opposed to the 335 as a standard calculation. In this paper, we describe the analysis of Malaysian wind direction data using the newly improved statistical software, ORIANA designed to analyze circular data. Exploratory data analysis (EDA) based on descriptive statistics, graphical display of the data and comparison of samples are discussed.