Big Data Affluence in Statistics Application: A Comparison of Real Life and Simulated Open Data

Abstract

Large data repositories or database management still remain a mirage and tough challenge to accomplish by most developing countries and establishments around the globe. This necessitates the need to improvise on the gathering of suitable data with a good spread to serve as a complement, in the absence of sufficient real-life data. Statisticians are increasingly posed with thought-provoking and even paradoxical questions, challenging our qualifications for entering the statistical paradises created by Big Data. Through classroom activities that involved both sourced real-life and simulated big data in R-environment, models were built and estimates obtained from the adopted techniques revealed the robustness of simulated datasets in a unified observation with improved significant values as reflected in the results. Students appreciated the use of such big data as it enhances their machine learning ability and the availability of sufficient data without delay.