**Acknowledgements**

not less important. The authors assume that this is mainly due to the procedure by which matrix *PR*,*<sup>C</sup>* was obtained for model EHS2. As could be expected, the smallest contribution to the EHS2 ensemble has its least precise member: the multiple linear regression (MLR). In **Figure 4**, a time series graph of the testing dataset with the observed flows and the flows simulated by the proposed ensemble model is seen. As can be seen, the predicted flows follow the real values with a high degree of precision, and the proposed ensemble approach could be

**Figure 4.** Time series of the testing dataset of the observed flows and the same flows simulated by the proposed ensemble

In this work, the authors deal with an investigation of the possible improvement of the river flow predictions. A new methodology was investigated in which ensemble modeling by datadriven models was applied and in which the harmony search was used to optimize the ensemble's structure. Because various data-driven models with strong prediction capability already exist, the authors were trying to evaluate in the case study presented in this paper (2-day ahead prediction of river flows), whether an ensemble paradigm would also bring some gain in cases when strong algorithms are used as ensemble members. Although the improvement in precision was not relatively as high as in the case when the ensemble consists of weak learners, it was proved that the ensemble model worked better than any of its constituents. These results mean, of course, that the proposed ensemble also works better than the ensembles with weak learners which are usually applied, because these were actually among the members of the proposed

used as an innovative alternative for flow predictions.

**4. Conclusion**

model in the year 1997.

166 Time Series Analysis and Applications

ensemble.

This work was supported by the Slovak Research and Development Agency under Contract No. APVV-15-0489 and by the Scientific Grant Agency of the Ministry of Education of the Slovak Republic and the Slovak Academy of Sciences, Grant No. 1/0665/15 and 1/0625/15.
