Search papers, labs, and topics across Lattice.
This paper critiques the prevalent use of mean squared error (MSE) in evaluating irregular time-series forecasting models, highlighting its bias due to timestamp sampling distributions. The authors introduce the Continuous-time Squared Error (CSE) metric, which utilizes importance weighting to provide a more accurate assessment of continuous-time predictive performance. Experimental results demonstrate that CSE outperforms MSE in recovering continuous-time risk across various datasets, suggesting that reliance on MSE may misrepresent model capabilities in real-world applications.
Relying on MSE for irregular time-series forecasting can lead to misleading evaluations, as it fails to account for timestamp sampling biases.
Existing research on irregular time-series forecasting has primarily focused on model design, while evaluation metrics remain insufficiently studied. Existing benchmarks typically use mean squared error (MSE) as the evaluation metric. We show that, in irregular forecasting, MSE is determined not only by the model prediction but also by the sample-specific timestamp sampling distributions, leading to a biased assessment of the models' continuous-time predictive performance. To address this issue, we propose the Continuous-time Squared Error (CSE), which employs importance weighting to eliminate the influence of the timestamp sampling distributions. We further theoretically prove that CSE's asymptotic estimation error with respect to continuous-time risk is no greater than that of MSE. Finally, we construct a systematic benchmark covering synthetic, semi-synthetic, and eight real-world datasets to validate the effectiveness of CSE and systematically evaluate models' continuous-time predictive performance. Experiments show that CSE can recover continuous-time risk more accurately than MSE, while relying solely on MSE may not fully reflect models' continuous-time predictive performance in real-world scenarios. Our code can be obtained at https://github.com/hnu-vis/ITS-Bench.
