Introduction

Time series measurements possess seasonality, autocorrelation, trend, and noise characteristics that need to be accounted during time series analysis. It should be noted that one or all of these characteristics can possessed by the time series dataset. Thus, it is significant to understand which data characteristics are inherent to the dataset under analysis. Autocorrelation - Tendency of current time series values to depend on previous values over time. Seasonality - Characteristic of the time series data to possess distinct repeated patterns over time. Trend - Tendency of time series data to go up, down or stay the same over time. Below are some of the approaches to examination of time series data characteristics using python.