What is the term for a modeling error when a model is too simplistic to capture data structures?

The term for a modeling error when a model is too simplistic to capture the underlying data structures is underfitting. Underfitting occurs when a model cannot adequately represent the complexities of the data, resulting in poor performance on both the training set and unseen data. This typically happens when the model has too few parameters or is based on overly simplistic assumptions, making it unable to learn the patterns and relationships present in the data.

In contrast, overfitting refers to a scenario where a model learns noise and random fluctuations in the training data instead of the actual underlying patterns, often resulting in poor generalization to new data. Noise refers to random variations and errors in the data that can obscure the true signals one is trying to model. Bias typically speaks to the tendencies of a model to make systematic errors in predictions due to simplified assumptions or limitations in the model design itself.

By recognizing that underfitting reflects the inability to capture the complexity of the data effectively, it becomes clear why this term is the most appropriate in the context of the question.