Method that improved area selection and consistently enhance forecast skill.

CGIAR Research Program on Climate Change, Agriculture and Food Security. 2017. Method that improved area selection and consistently enhance forecast skill. Reported in Climate Change, Agriculture and Food Security Annual Report 2017. Innovations.

Two statistical models were used to evaluate the seasonal forecasts: Canonical Correlation Analysis (CCA) implemented in CPT, and Multiple Factorial Analysis (MFA), which allows integrating several regions or several predictors simultaneously. Using the results from the teleconnection analysis, forecast skill in Colombia for the period form 1982-2013 was assessed.