Supplementary material to "Simple Box-Cox probabilistic models for hourly streamflow predictions"

This supplementary material details the calculation of the best-metric residual error model and the analysis of statistical significance of differences in the performance metrics of the competing residual error models.We look at performance rankings of the residual error models.We use statistical tests to assess whether differences in the performance metrics of competing residual error models are 15 significant.For example, to compare the reliability (R) of three residual error models A, B, C and D.