A Markov chain approach to crop yield forecasting Academic Article uri icon

abstract

  • This study proposes a methodology for forecasting crop yields at intermediate times in the growing season using Markov chain theory. A Markov chain is constructed, based on historical data, to provide forecast distributions of crop yield for various crop and soil moisture condition classes at selected times prior to harvest. Expected yield and the associated standard error are obtained for each condition class. The methodology is compared to a regression approach in which the independent variables are the various crop and soil moisture conditions. The Markov chain approach requires less stringent assumptions and provides more information than the regression approach. However, the potential loss of precision in the forecast using this approach requires separate evaluation for each application. A data base created by the CERES-Maize model, which simulates the growth and development of a corn crop, is used to demonstrate the development of the forecast yield distributions using the Markov chain approach. © 1985.

author list (cited authors)

  • Matis, J. H., Saito, T., Grant, W. E., Iwig, W. C., & Ritchie, J. T.

citation count

  • 14

publication date

  • January 1985