Backward elimination is a statistical method used in model selection, particularly in regression analysis. It starts with a full model that includes all potential predictors and systematically removes the least significant variables one at a time. The goal is to improve the model's performance by retaining only the most relevant predictors. This process continues until all remaining variables contribute significantly to the model, as determined by a chosen criterion, such as the p-value. Backward elimination helps simplify models, making them easier to interpret while maintaining predictive accuracy.