Abstract: A distribution of values of time-series data is obtained. Based on the distribution of the values, the time-series data is sampled to generate an anomaly preserving version of the time-series data. Via a trained machine learning model, a reconstructed version of the time-series data is generated based on the anomaly preserving version of the time-series data.

Abstract: Systems and methods for managing an asset portfolio. A system generates a detailed trading schedule that converts a current portfolio into a desired portfolio. The schedule is generated using machine learning and is based on a number of inputs including the current portfolio, a desired portfolio, an execution timeline, as well as user supplied constraints. Once generated, the system evaluates the schedule using one or more market models to determine if the schedule will be feasible given market reactions based on the one or more models. The system iterates the generation/evaluation loop until the best possible schedule is arrived at. In addition, the system may provide recommendations for not only brokers to be used when executing the trades but also trading algorithms that the brokers may use when implementing the schedule.