Stationarity Estimation for Time Series Rainfall Data

Abstract: In this paper, fractality and stationarity of average rainfall value of India have been investigated with a focus on the self similarity pattern and nature of frequency oscillation of the two parameters namely the average rainfall value of North India and South India. The time series of these parameters in between 31st Oct 2006 to 1st Nov 2017, of North India and South India have been chosen for extracting the nature of scaling (fractality) and stationary behavior using statistical methodologies. Hurst Exponent for the time series have been calculated using the methodologies like Higuchi Fractal Dimension (HFD) and General Hurst Estimation (GHE). It has been noticed that both the time series show Short Range Dependent (SRD) anti-persistent behavior. Continuous Wavelet Transform (CWT) method has been incorporated to identify the stationarity/non-stationarity of the data-series where both the time series exhibit the non stationarity.

Keywords: Rain Fall Data; Hurst Exponent; Generalized Hurst Estimation (GHE); Higuchi’s Fractal Dimension (HFD); CWT