Using randomness compensating factors to improve forecast accuracy

A computer-implemented method comprising:

• receiving input data that includes a plurality of observations made at corresponding time slices;
• provide the input data to one or more time-based prediction models that predict one or more future values for one or more future time slices;
• determine one or more randomness compensating factors, the one or more randomness compensating factors correspond to one or more features indicative of one or more events that occur on one or more of the corresponding time slices;
• determine, for one or more of the corresponding time slices, one or more patterns of observations associated with the one or more randomness compensating factors;
• based at least in part on the one or more randomness compensating factors and the one or more patterns, modify the prediction of the one or more time-based prediction models;
• and
• in response to the modifying of the prediction, presenting, to a user computer device, an indication associated with the modified prediction.

A computer readable storage medium having program instructions embodied therewith, the program instructions executable by one or more processors to cause the one or more processors to:

• receive input data that includes one or more observations made at one or more corresponding time slices;
• provide the input data to one or more time-based prediction models that predict one or more values for one or more future time slices;
• determine one or more randomness compensating factors, the one or more randomness compensating factors correspond to one or more features indicative of one or more events associated with the one or more corresponding time slices;
• based at least in part on the determining of the one or more randomness compensating factors and the prediction of the one or more time-based prediction models, predict a second one or more values for at least one of the one or more future time slices;
• and
• in response to the predicting of the second one or more values, presenting, to a user computer device, an indication associated with the predicted second one or more values.