Trend Analysis and Equivalence testing for environmental data
TimeTrends was developed to make life easier for people who carry out trend analyses, equivalence tests or develop multivariate models by providing a series of linked procedures to:
- Import data/time data easily
- display time trends with smoothing routines applied (Lowess,Gams)
- statistical analysis by season (Seasonal Kendall test)
- box and whisker plots by season with Kruskal-Wallis test
- annual and seasonal event analysis
- adjustment of data for flow variations (Lowess, Log-log, Gams)
- Parametric trend analysis with equivalence testing of slope.
- Equivalence testing of means
- determine trends through multivariate models (linear regression, generalised additive (GAM), or Loess)
- piecewise regression to aid interpretation of trend and CUSUM plots.
The steps in a typical analysis are:
- open file of environmental data
- plot variable against time and fit linear and smoothing functions
- statistical Kendall analysis of trends with time and season
- seasonal boxplots and event analysis
- multivariate analysis: regression, GAM or Loess
Data and results of analyses are shown on a series of pages with tabs.
Most analyses have both graphical and text outputs, with the results displayed on a graph page and text output on the results page.
Download TimeTrends 32 bit (16 Mb) version 11