Online Forecasting Matrix Factorization

TL;DR

This paper introduces online matrix factorization methods for forecasting high-dimensional time series with missing data, enabling real-time predictions through low-rank embeddings and autoregressive modeling.

Contribution

It presents novel online matrix factorization techniques and a recursive estimator for effective real-time forecasting of streaming high-dimensional data.

Findings

Effective in handling missing data

Scalable to large datasets with tens of millions of measurements

Improves forecasting accuracy in real-world scenarios

Abstract

In this paper the problem of forecasting high dimensional time series is considered. Such time series can be modeled as matrices where each column denotes a measurement. In addition, when missing values are present, low rank matrix factorization approaches are suitable for predicting future values. This paper formally defines and analyzes the forecasting problem in the online setting, i.e. where the data arrives as a stream and only a single pass is allowed. We present and analyze novel matrix factorization techniques which can learn low-dimensional embeddings effectively in an online manner. Based on these embeddings a recursive minimum mean square error estimator is derived, which learns an autoregressive model on them. Experiments with two real datasets with tens of millions of measurements show the benefits of the proposed approach.