Using Linear Regression to Predict Binary Price Direction
Summary
The document raises a modeling question about predicting whether a future price move is positive using a label that takes only two values. The label compares a later close with an earlier open and assigns one class for an increase and the other for a non-increase. It asks whether ordinary linear regression is suitable for training on this target and whether its predictions remain useful.
No answer, experiment, or performance evidence is included, so the document does not establish that linear regression is appropriate or that its outputs are calibrated probabilities. It is best read as a prompt about choosing a model for binary classification and interpreting continuous regression outputs. The discussion also omits details such as feature definitions, validation design, class balance, and evaluation metrics, all of which would be needed to judge practical usefulness.
Key ideas
- The target labels whether a future price comparison is positive or not.
- The document questions whether linear regression is appropriate for a binary target.
- It provides no experiment or answer establishing the usefulness of the resulting predictions.
- Model evaluation would require details about features, validation, and suitable metrics.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.