Indian Forum for Water Adroit

Critical Values for Sen’s Trend Analysis

Pankaj Dey

  • *****
  • Thanked: 110 times
  • +112/-0
    • View Profile
  • Institute : Indian Institute of Science
  • Programming language : MATLAB, R
Critical Values for Sen’s Trend Analysis
« on: August 31, 2018, 03:26:54 PM »

Trends in measured hydrologic data can greatly influence projected values; therefore, trends need to be identified and quantitatively modeled. First, it is necessary to verify whether or not a trend actually exists in the sample data. Tests of significance are most often used to verify whether or not a trend is statistically significant. Şen provided an easy-to-apply method of identifying the presence of trends in time series, but did not provide a quantitative method of verifying the statistical likelihood of an assumed trend in a measured time series [Şen, Z. 2012. “Innovative trend analysis methodology.” J. Hydrol. Eng. 17 (9): 1042–1046]. A method of quantifying Sen’s approach is developed herein, with critical values of the test statistic developed to provide a means of making objective decisions. Critical values are presented for rejection probabilities from 10% to 0.1%. The power of the test, which is similar to that of other trend tests, is also approximated; analyses indicate powers from 10% to 30% for small samples.

The following users thanked this post: Sonali