# Marginalize, Rather than Impute: Probabilistic Wind Power Forecasting with Incomplete Data

arXiv: https://arxiv.org/abs/2403.03631 · HTML: https://mlsanity.com/paper/2403.03631/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 45 references that are in the map.
- First posted on arXiv 6 Mar 2024.

## Builds on (most cited first, 13 of 13)

- [Variational Inference with Normalizing Flows](https://mlsanity.com/paper/1505.05770/) (2015)
- [Normalizing Flows for Probabilistic Modeling and Inference](https://mlsanity.com/paper/1912.02762/) (2019)
- [DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks](https://mlsanity.com/paper/1704.04110/) (2017)
- [Importance Weighted Autoencoders](https://mlsanity.com/paper/1509.00519/) (2015)
- [On the consistency of supervised learning with missing values](https://mlsanity.com/paper/1902.06931/) (2019)
- [Regression markets and application to energy forecasting](https://mlsanity.com/paper/2110.03633/) (2021)
- [Continuous and Distribution-free Probabilistic Wind Power Forecasting: A Conditional Normalizing Flow Approach](https://mlsanity.com/paper/2206.02433/) (2022)
- [Resilient Neural Forecasting Systems](https://mlsanity.com/paper/2203.08492/) (2022)
- [Wind energy forecasting with missing values within a fully conditional specification framework](https://mlsanity.com/paper/2203.08252/) (2022)
- [Availability Adversarial Attack and Countermeasures for Deep Learning-based Load Forecasting](https://mlsanity.com/paper/2301.01832/) (2023)
- [Cyber Risk Analysis of Combined Data Attacks Against Power System State Estimation](https://mlsanity.com/paper/1708.08349/) (2017)
- [Probabilistic wind power forecasting resilient to missing values: an adaptive quantile regression approach](https://mlsanity.com/paper/2305.14662/) (2023)
- [Improving Variational Autoencoder Estimation from Incomplete Data with Mixture Variational Families](https://mlsanity.com/paper/2403.03069/) (2024)

## Abstract (arXiv)

Machine learning methods are widely and successfully used for probabilistic wind power forecasting, yet the pervasive issue of missing values (e.g., due to sensor faults or communication outages) has received limited attention. The prevailing practice is impute-then-predict, but conditioning on point imputations biases parameter estimates and fails to propagate uncertainty from missing features. Our approach treats missing features and forecast targets uniformly: we learn a joint generative model of features and targets from incomplete data and, at operational deployment, condition on the observed features and marginalize the unobserved ones to produce forecasts. This imputation-free procedure avoids error introduced by imputation and preserves uncertainty aroused from missing features. In experiments, it improves forecast quality in terms of continuous ranked probability score relative to impute-then-predict baselines while incurring substantially lower computational cost than common alternatives.