This one is for statistics.<p>I have found the explanation of statistical concepts through the lens of linear algebra immensely intuitive. A simple, short and clear illustration of this is in 'The Geometry of Multivariate Statistics' by Thomas D. Wickens, which I purchased solely based on it's title. It goes through the geometric interpretation of univariate, and multivariate linear regression, then goes into the geometric interpretation of correlation, collinearity impact on prediction, PCAs, and statistical tests. Warning: This book assumes you have some very basic statistical background.<p>Funnily enough, recently I've been going through Strang's 'Introduction to Linear Algebra' textbook, and he also goes through derivation of mulitvariate statistics in the same fashion. I like the way he builds up the geometric interpretation of regression by building up from a exploration of column/row spaces, orthogonality, projection matrices, and from there, seamlessly introduces solving the LLS as a problem that can be solved with a projection matrix. That being said, I find Wicken does a better job of illustrating his concepts, which is most intuitive modality to interpret this.