gdp-nowcasting-dynamic-factorlisted

Nowcast the quarter you are in from monthly data with a ragged edge, using statsmodels' DynamicFactorMQ - and score it against the benchmarks it has to beat. TRIGGER - nowcast, nowcasting, GDPNow, Atlanta Fed GDP tracker, New York Fed Staff Nowcast, DynamicFactorMQ, endog_quarterly, k_endog_monthly, fit_em, dynamic factor model, mixed frequency, monthly and quarterly in one model, ragged edge, jagged edge, unbalanced panel, Mariano-Murasawa, Banbura Modugno, bridge equation, MIDAS, "how do I combine monthly indicators into a GDP forecast", news decomposition of a data release. SKIP for the Kalman filter and smoother themselves (state-space-and-kalman), for univariate forecasting and its baselines (time-series-forecasting-models), for release timestamps (macro-release-calendar-and-embargo), and for vintages of the inputs (real-time-macro-backtesting).
howard-lynn-ye/fin-skills · ★ 1 · AI & Automation · score 77

Install: claude install-skill howard-lynn-ye/fin-skills

# GDP nowcasting with a dynamic factor model **A nowcast is not a forecast of the future — it is an estimate of a number that already happened and has not been published yet.** Everything hard about it is in the shape of the data: monthly indicators arriving on different days with different lags, a quarterly target observed once every three months, and a bottom edge of the panel that is jagged rather than flat. The model that handles that is a state-space one, and it is already installed. The Kalman machinery itself lives in `../../../fin-models/skills/state-space-and-kalman/SKILL.md`; univariate baselines and how to score a forecast live in `../../../fin-models/skills/time-series-forecasting-models/SKILL.md`. This skill is the mixed-frequency, ragged-edge part and the benchmark discipline. Everything marked ✅ Measured is printed by `scripts/nowcast.py` (numpy + pandas + scipy, seed 20260909, **6–25 s** depending on machine load, no network). Everything marked ✅ source-verified was read at the URL or in the installed package on 2026-09-09. ## 1. ✅ `DynamicFactorMQ` is the NY-Fed-shaped model, already in a BSD dependency ✅ source-verified in `statsmodels/tsa/statespace/dynamic_factor_mq.py`, statsmodels 0.15.0, read in the installed package. The class docstring opens: > *"Implementation of the dynamic factor model of Bańbura and Modugno (2014) and Bańbura, > Giannone, and Reichlin (2011). Uses the **EM algorithm** for parameter fitting… Can > incorporate **monthly/quarte