When statisticians talk about linear models, they talk about the parameters being linear, not your variables x_0..x_n. So y = a*sin(x) + b is a linear model, be

When statisticians talk about linear models, they talk about the parameters being linear, not your variables x_0..x_n. So y = a*sin(x) + b is a linear model, because y is linear in a and b.