314: Double cross-validated

The double cross-validated calculation option fits a CCA , PCR , or MLR model using all data within the training period. It also works with the GCM option. The option produces double cross-validated forecasts for each year in the training period. Deterministic and probabilistic double cross-validated forecasts are made.

At each double cross-validation step one year is omitted from the training period, and regular cross-validation is performed on the remaining part of the training period before making a prediction for the year omitted in the double cross-validation. This process is repeated with the next year omitted until a doubly cross-validated prediction has been made for each year in the training period.

The double cross-validated forecasts are made available for output to a file.

Last modified: