I like to remember it by thinking about what does “orthogonal” mean in geometry. Two lines are orthogonal if they have are 90 degrees from each other in euclidian (x and y) space.
The way you represent variables in the math of statistics is in a surface of multiple variables (let’s just think of two to make it simple). If I can chart one variable along the x axis and one on the y axis, then the math becomes much simpler because I don’t have to take into account the relationship between the two. That means your variables are orthogonal and you can simplify the equations.
Latest Answers