Speaker
Sergey Kunin-Bogoiavlenskii
Description
Aggregating forecasts from multiple experts is a valuable method to improve prediction accuracy. Our work examines the influence of hyperparameters on the accuracy of the aggregation algorithm for a countable number of experts. We implement a time series generator with specified properties and an aggregating forecasting model. We conduct a series of experiments with various hyperparameters of the algorithm. The experiments confirm that these hyperparameters have a significant influence on the algorithm's performance.
Primary authors
Mrs
Anastasia Zukhba
(доцент кафедры МОУ МФТИ)
Sergey Kunin-Bogoiavlenskii
