Sas Institute Inc.

Yung-Hsin Chien

Yue Li

Pu Wang

Jingrui Xie

Ann Mary McGuirk

Mahesh V Joshi

Michele Angelo Trovero

Yongqiao Xiao

Jerzy Michal Brzezicki

Phillip Mark Helmkamp

Thomas Burkhardt

Sagar Arun Mainkar

Steven Christopher Mills

Macklin Carter Frazier

Ron Travis Hodgin

Timothy Patrick Haley

Bhupendra Suresh Bendale

Necati Burak Meric

Randy Thomas Solomonson

Neha Bindumadhav Kulkarni

Sangmin Kim

A pipeline system for time-series data forecasting using a distributed computing environment is disclosed herein. In one example, a pipeline for forecasting time series is generated. The pipeline represents a sequence of operations for processing the time series to produce modeling results such as forecasts of the time series. The pipeline includes a segmentation operation for categorizing the time series into multiple demand classes based on demand characteristics of the time series. The pipeline also includes multiple sub-pipelines corresponding to the multiple demand classes. Each of the sub-pipelines applies a model strategy to the time series in the corresponding demand class. The model strategy is selected from multiple candidate model strategies based on predetermined relationships between the demand classes and the candidate model strategies. The pipeline is executed to determine the modeling results for the time series.

Demand classification based pipeline system for time-series data forecasting

A hierarchical structure (e.g., a hierarchy) for use in hierarchical analysis (e.g., hierarchical forecasting) of timestamped data can be automatically generated. This automated approach to determining a hierarchical structure involves identifying attributes of the timestamped data, clustering the timestamped data to select attributes for the hierarchy, ordering the attributes to achieve a recommended hierarchical order, and optionally modifying the hierarchical order based on user input. Through the approach disclosed herein, a hierarchy can be generated that is designed to perform well under hierarchical models. This recommended hierarchy for use in hierarchical analysis may be agnostic to any planned hierarchy provided by or used by a user to otherwise interpret the timestamped data.

Computer implemented systems for automatic hierarchy for large scale time series data sets

Systems and methods are provided for performing data mining and statistical learning techniques on a big data set. More specifically, systems and methods are provided for linear regression using safe screening techniques. Techniques may include receiving a plurality of time series included in a prediction hierarchy for performing statistical learning to develop an improved prediction hierarchy. It may include pre-processing data associated with each of the plurality of time series, wherein the pre-processing includes tasks performed in parallel using a grid-enabled computing environment. For each time series, the system may determine a classification for the individual time series, a pattern group for the individual time series, and a level of the prediction hierarchy at which the each individual time series comprises an need output amount greater than a threshold amount. The computing system may generate an additional prediction hierarchy using the first prediction hierarchy, the classification, the pattern group, and the level.

Improving accuracy of predictions using seasonal relationships of time series data

In some examples, a processing device can receive prediction data representing a prediction. The processing device can also receive files defining abnormal data-point patterns to be identified in the prediction data. The processing device can identify at least one abnormal data-point pattern in the prediction data by executing customizable program-code in the files. The processing device can determine an override process that corresponds to the at least one abnormal data-point pattern in response to identifying the at least one abnormal data-point pattern in the prediction data. The processing device can execute the override process to generate a corrected version of the prediction data. The processing device can then adjust one or more computer parameters based on the corrected version of the prediction data.

Predicting and adjusting computer functionality to avoid failures

Systems and methods for machine learning using classifying, clustering, and grouping time series data

Computer-implemented systems and methods are provided to perform accuracy analysis with respect to forecasting models, wherein the forecasting models provide predictions based upon a pool of production data. As an example, a forecast accuracy monitoring system is provided to monitor the accuracy of the forecasting models over time based upon the pool of production data. A forecast model construction system builds and rebuilds the forecasting models based upon the pool of production data.

System and methods for retail forecasting utilizing forecast model accuracy criteria, holdout samples and marketing mix data

Computer-implemented systems and methods generate forecasts or estimates with respect to one or more attributes contained in an attribute-based hierarchy. Physical hierarchical data and attribute input data are received so that an attribute-based hierarchy can be created. A mapping table is created that indicates relationships between the attribute-based hierarchy and the physical hierarchy, wherein the attribute-based hierarchy is accessed during model forecasting analysis or model estimation analysis.

Attribute based hierarchy management for estimation and forecasting

Systems and methods for providing estimations for a product for purchase at a plurality of stores. Groups of stores are generated based upon similarity of store demand data. For each group, a distribution is determined with respect to the attribute of the product. The distribution is used to provide estimations with respect to the product to be provided at the stores.

Computer-implemented systems and methods for product attribute estimations

Computer-implemented systems and methods for determining demand of products. A system and method can be configured to determine a price with respect to a first attribute of a first product. This determination is based upon the price data of the products which compete with the first product and whose attributes are alike with respect to the first product's attributes except for a first attribute. The determined single price is used in a mathematical model for determining demand for the first product.

Growing community of inventors

Apex, NC, United States of America

Yung-Hsin Chien