AI Paper Atlas

Research map / Time series / Causal and generative time series

MixSeq: Connecting Macroscopic Time Series Forecasting with Microscopic Time Series Data

Zhibo Zhu, Ziqi Liu, Ge Jin, Zhiqiang Zhang, Lei Chen, JUN ZHOU, Jianyong Zhou

NeurIPS 2021 · Time series › Causal and generative time series · approach: Robust forecasting and evaluation · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Cluster microscopic series with MixSeq, then forecast and aggregate them instead of forecasting the macro series directlyConvTrans on Rossmann · RossmannSMAPE: 0.1028(0.0091) -> 0.0961(0.0019)Improved
Cluster microscopic series with MixSeq, then forecast and aggregate themConvTrans on M5 · M5SMAPE: 0.0245(0.0024) -> 0.0227(0.0006)Improved
Cluster microscopic series with MixSeq, then forecast and aggregate themConvTrans on Wiki · WikiSMAPE: 0.0466(0.0001) -> 0.0440(0.0010)Improved
Use MixSeq clustering rather than macro forecastingConvTrans on Rossmann · RossmannR0.5 loss: 0.1861 -> 0.1847Improved
Use MixSeq clustering rather than macro forecastingConvTrans on Rossmann · RossmannR0.9 loss: 0.0822 -> 0.0813Improved
Use MixSeq clustering rather than the strongest listed alternativeConvTrans on M5 · M5R0.5 loss: 0.0497 -> 0.0460Improved
Use MixSeq clustering rather than the strongest listed alternativeConvTrans on M5 · M5R0.9 loss: 0.0257 -> 0.0238Improved
Use MixSeq clustering rather than the strongest listed alternativeConvTrans on Wiki · WikiR0.5 loss: 0.0961 -> 0.0901Improved
Use MixSeq clustering rather than the strongest listed alternativeConvTrans on Wiki · WikiR0.9 loss: 0.0531 -> 0.0516Improved
Use a flexible Seq2seq mixture model rather than an ARMA mixtureSynthetic DeepAR time-series clustering · DeepAR synthetic data, 2 clustersRand Index: 0.7995(0.2734) -> 0.9986(0.0003)Improved
Aggregate time series by their known generating component before forecastingForecasting macroscopic series in toy mixtures · GP and ARMA mixtures, 3 and 5 clustersSMAPE: 0.0263 -> 0.0210; 0.0242 -> 0.0198; 0.5870 -> 0.3590; 0.5940 -> 0.3840Improved
Use 3 clusters, matching the ground-truth cluster count, rather than 2 (ablation)MixSeq with ARMA forecasting · Synthetic ARMA forecastingSMAPE: 0.774 -> 0.731Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy