quadratic discriminant analysis

Quadratic Discriminant Analysis (QDA) is a statistical method used for classifying objects into categories based on their features. It assumes each category has its own unique pattern of variation and spread, represented mathematically by a quadratic (curved) boundary. QDA models these differences by estimating the probability of an object belonging to each category, considering how features vary together within each group. When a new object is tested, QDA compares these probabilities and assigns it to the category with the highest likelihood, allowing for flexible, non-linear distinctions between groups.