Regression vector compressed feature vector machines for forecasting use of timed stocks

摘要

The present disclosure includes a system for regression tree modified feature vector machine learning models for usage forecasting in timed inventory. An on-line computing system receives the feature vectors for the listing and inputs the feature vectors and the modified feature vectors into a demand function to generate a demand estimate. The system inputs demand estimates into the likelihood model to generate a set of demand likelihoods, each demand likelihood receives a trade request at each of the set of timed inventory and test price and test time until expiration. Represents the likelihood of The system further trains a regression tree model based on a set of training data, the set generating a demand likelihood from the set and a demand estimate used to generate the demand likelihood. Each of the test price used and the test time period until expiration is provided.