Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Parametric Prior Mapping Framework for Non-stationary Probabilistic Time Series Forecasting

About

Effectively modeling non-stationary dynamics in probabilistic multivariate time series(MTS) forecasting requires balancing expressiveness with robustness. Existing parametric approaches benefit from strong inductive biases but lack flexibility, whereas deep generative models struggle to capture complex temporal dependencies without extensive data and computation. We introduce Parametric Prior Mapping (PPM), a framework that injects parametric structural priors into a generative modeling process. Specifically, PPM utilizes a parametric estimator to derive a dynamic, adaptive prior that guides the learning of a complex predictive distribution via a learnable mapping. This design allows the model to retain the efficiency of parametric methods while exploiting the expressive power of generative models. Trained with a hybrid objective, PPM yields precise forecasts with well-calibrated uncertainty estimates. Empirical results show that PPM outperforms existing baselines in handling non-stationary data, offering a superior trade-off between accuracy and computational efficiency. The code is available at https://github.com/ljl8336/PPM.

Jinglin Li, Jun Tan, QI Fang, Ning Gui• 2026

Related benchmarks

TaskDatasetResultRank
Time Series ForecastingETTm2
MSE0.247
536
Time Series ForecastingElectricity
MSE0.182
237
Time Series ForecastingETTh2
MSE0.376
88
Time Series ForecastingWeather
MSE0.242
55
Probabilistic time series forecastingETTh1
CRPS0.337
21
Probabilistic time series forecastingETT m1
CRPS0.314
15
Probabilistic time series forecastingETTm2
CRPS0.237
15
Probabilistic time series forecastingETTh2
CRPS0.306
15
Probabilistic time series forecastingElectricity
CRPS0.206
15
Probabilistic time series forecastingWeather
CRPS0.215
8
Showing 10 of 11 rows

Other info

Follow for update