XMatch improves time series forecasts using adaptive exogenous pattern matching
XMatch: Enhancing Covariate-Aware Time Series Forecasting through Tree-Structured Exogenous Matching
Machine Learning
Summary
Forecasting is about guessing future numbers based on past data. Sometimes, outside factors (called exogenous variables) help with these guesses, but their effects can be tricky to figure out. The authors show that certain outside conditions usually lead to only a few typical future patterns in the data being predicted. So, they built a tool that looks for past times with similar outside conditions and uses what actually happened then to make better predictions. This method adjusts how many outside factors it looks at based on how closely things match and how much past data it has, helping it forecast more accurately.
What this means in practice
- •For energy demand planners: Improve electricity demand forecasts by matching upcoming weather and calendar patterns to past similar conditions for more accurate load predictions.
- •For inventory managers: Enhance sales forecasts by using future promotional and seasonal variables to find comparable past sales scenarios, improving stock planning.
Authors
Ziyang Zhang, Hanyin Cheng, Xiangfei Qiu, Yang Shu, Bin Yang, Chenjuan Guo
Abstract
Future exogenous variables provide valuable information for forecasting endogenous time series. Existing covariate-aware methods primarily learn the direct influence of exogenous variables on endogenous variables. However, these effects can be complex and change with the pattern of the exogenous variables, making them difficult to capture. Beyond this perspective, we observe that a given exogenous pattern often co-occurs with only a small set of endogenous response patterns. These associations motivate a strategy that matches future and historical exogenous patterns and uses the corresponding endogenous patterns to enhance forecasting. However, in real-world forecasting scenarios with multiple exogenous variables, each exogenous variable provides a distinct dimension for matching, creating a dilemma for this strategy between precise matching and sufficient historical support. To bridge this gap, we propose XMatch (EXogenous MATCHing), a covariate-aware forecasting model that realizes the aforementioned strategy through a tree-structured matching process that adaptively adjusts the number of exogenous variables used as matching conditions. Specifically, we first introduce the ProtoTree Creator, which organizes historical correspondences between exogenous and endogenous patterns into a ProtoTree, whose deeper levels incorporate additional exogenous variables for matching. For forecasting, we then design the ProtoTree Matcher, which uses future exogenous variables to query the ProtoTree and adaptively determines how many exogenous variables to use for matching based on exogenous pattern similarity and historical support. Finally, the matched endogenous patterns are used as explicit historical evidence to enhance forecasting. Extensive experiments on 12 real-world datasets demonstrate that XMatch outperforms state-of-the-art baselines.