Function fit_and_rss

fn fit_and_rss(
    y: &[f64],
    predictors: &[&[f64]],
    lag: usize,
    n_obs: usize,
) -> f64

Fit a simple autoregressive model and return residual sum of squares.

y[lag..] = sum over each series in predictors of (sum_l beta_l * series[t-l]) + epsilon

Uses the normal equation (X^T X)^{-1} X^T y via iterative least squares simplified to a simple approach.