nx_forecast.nx

nx_forecast.nx -- time-series forecasting primitives. Classical recursive forecasters complementing nx_kalman: 1. Simple Exponential Smoothing (Brown 1956) s_t = alpha * x_t + (1 - alpha) * s_{t-1} 2. Holt Double Exponential (Holt 1957) l_t = alpha * x_t + (1 - alpha) * (l_{t-1} + b_{t-1}) b_t = beta * (l_t - l_{t-1}) + (1 - beta) * b_{t-1} forecast(h) = l_t + h * b_t 3. Autoregressive AR(p) forecaster x_t = phi_1 x_{t-1} + ... + phi_p x_{t-p} All smoothing parameters in Q14 fixed-point. Pure i64. What this unlocks: + sensor smoothing without state-model (lighter than Kalman) + trend extrapolation + AR-model rollout for short-horizon prediction + anomaly detection (deviation from forecast) Sibling of nx_ts.nx (which handles Unix-epoch <-> calendar); this module forecasts numeric time series. genealogy_id: brown_1956_ses + holt_1957_double_exp + box_jenkins_1970 lineage_id: recursive_exponential_smoother + autoregressive_predictor