Skip to content
Predict.ai

Data & drivers

Covariate

Also called exogenous variable

A covariate is an extra series the forecast is allowed to see besides the target — price, promo, weather, spend. It only helps if you will have it (or a forecast of it) at prediction time.

The target's past explains a lot. Covariates explain the leftovers: why this Thursday was not like last Thursday. A promo flag with depth will do more than twenty vague 'campaign' tags.

Future covariates are a contract. If you use tomorrow's actual temperature, you cheated. If you use a weather forecast available this morning, you played fair. Write that down or the next person will cheat by accident.

More columns are not more wisdom. Each extra series is a chance to overfit last quarter's accident. Add them one at a time and keep the ones that move the backtest.

A two-day lead

Leading indicator lagclicksorderslag 2
Clicks (top) peak two days before orders (bottom). Same-day correlation would call them twins. The lag is the actual product.

Read next

Related terms

Try it on your data.

Connect an outcome and its history. Predict.ai finds the drivers, races the models, and keeps the forecast live — in the workspace, over the API, and through agents.