MATLAB tool for calculating and plotting goodness-of-fit indicators considering measurement uncertainty and/or model uncertainty Conference Paper uri icon

abstract

  • Measurement uncertainty and model uncertainty should be accounted for in model application and evaluation. By modifying the error term in pair-wise comparisons of measured and predicted values for goodness-of-fit indicators, methods have been proposed to include measurement uncertainty and model uncertainty in these calculations. This paper presents a process-oriented MATLAB program to calculate goodness-of-fit indicators that includes these uncertainty estimates. The program requires minimum input files and user specified parameters. Nine types of common probability distribution combinations for hydrologic and water quality data can be selected. The program outputs include input data statistical properties, input data plots, and goodness-of-fit indicator values. The developed MATLAB tool, Goodness of Fit Indicator Tool (GOFIT), will be available for users as a stand-alone application or may also be integrated into other tools used in model optimization.

published proceedings

  • American Society of Agricultural and Biological Engineers Annual International Meeting 2009, ASABE 2009

author list (cited authors)

  • Dong, L., Migliaccio, K. W., Harmel, R. D., & Smith, P. K.

publication date

  • January 2009