howard-lynn-ye
- 103 skills
- 0 followers
- 1 day ago last updated
- ▌ Macro Release Calendar And Embargo · howard-lynn-yeBuild the timestamp at which a macro number becomes tradeable - release date, clock time, timezone - and know where the release mechanics changed under your sample. TRIGGER - release calendar, economic calendar, release date vs reference date, available_at, as-of join on a macro series, "when was this number published", 8:30 ET, embargo, press lock-up, media lockup, pre-release access, WASDE noon, EIA Wednesday 10:30, natural gas storage Thursday, holiday release schedule, "why is my macro feature one day early", forward-fill a monthly series onto daily bars, DST offset on a release timestamp, event-study window around a print. SKIP for vintages and revisions to the value itself (real-time-macro-backtesting), for where the series live (fundamental-and-macro-data), for exchange sessions and holidays (us-market-rules), and for measuring fills you already have (execution-cost-analysis).
- ▌ Alpha Combination And Neutralization · howard-lynn-yeScore several alphas, combine them, and strip the exposures you did not mean to take. TRIGGER - information coefficient, IC, rank IC, ICIR, IC decay, "is my IC good", IC t-stat, Newey-West on IC, overlapping forward returns; combining alphas, blending signals, alpha weighting, z-score or rank combination, multi-factor signal; sector neutral, beta neutral, market neutral signal, industry neutralization, residualize the alpha, cross-sectional regression residuals, orthogonalize signals; winsorize, clip outliers, cross-sectional standardization; turnover-aware combination, signal smoothing, "my alpha dies after costs". SKIP for one time-series trend signal (trend-following-models), for the factor library and the alphalens forward-return convention (factor-and-timeseries-research, lib-alphalens), for weights under constraints (portfolio-and-risk), and for whether the survivor is real (backtest-validation).
- ▌ Macro Regime And Recession Indicators · howard-lynn-yeRecession probabilities, the Sahm rule and yield-curve inversion - and the fact that the NBER label they are all scored against was assigned years after the fact. TRIGGER - USREC, NBER recession dates, recession indicator, recession probability, recession dummy, "recession-aware" strategy, regime flag from FRED, RECPROUSM156N, Sahm rule, SAHMREALTIME, SAHMCURRENT, "unemployment rose 0.5 points", yield curve inversion, T10Y3M, T10Y2Y, 2s10s, inverted curve recession signal, "how long after inversion", business cycle dating, "when did NBER announce", labelling recessions for a classifier. SKIP for fitting HMMs and the smoothed-versus-filtered timing of estimated regimes (regime-detection), for the vintage A/B on a macro strategy (real-time-macro-backtesting), for nowcasting GDP itself (gdp-nowcasting-dynamic-factor), and for where the underlying series live (fundamental-and-macro-data).