f/fundlab
Greg Pomerantz f68b239b9a Factor screen (v2, 35 drivers) + return-driver clusters
factors.py: full OLS of all 2,384 funds on a 35-driver basis
(overinclusive, no portfolio-corr screening - corr is a replacement
signal, not a rejection). v1's 21 + lqd/hyg/prefs/emb/tip/shy/vtv/
8 sectors/CTA/commodities. Basis fixes: drop vea/vug (dupes of
efa/qqq), drop finux (TERMINATED 2017 - silently zeroed the
complete-case mask; v1 forward selection never hit this), drop bil
(shv/bil near-null -> offsetting shv+64/bil-54 noise fits),
residualize vblix on ivv+tlt (pure vol axis), ridge 0.02.

cluster.py: hierarchical tree saved but fixed-k cuts degenerate
(most funds are blends -> one 2250-fund blob); k-means++ (deterministic)
is the useful grouping, re-run live in-app for any k.

app: Return-driver clusters expander (k slider 10-60, summary table,
member table sorted by alpha-t). Findings at k=30 in RESEARCH.md.
2026-08-27 08:35:42 -04:00
..
nport_cache Fund Lab: returns-based strategy decomposition for the 16 shortlist funds 2026-08-26 11:59:59 -04:00
pool Fund Lab: returns-based strategy decomposition for the 16 shortlist funds 2026-08-26 11:59:59 -04:00
__init__.py fundlab: EDGAR investment-objective fetcher + curated fallback 2026-08-25 18:15:48 -04:00
cluster_kmeans.json Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
cluster_tree.json Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
cluster.py Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
dbmine.py Find candidate funds NOT in the DB: exhaustive EDGAR 497-universe pass 2026-08-26 15:22:52 -04:00
decompose_results.json Fund Lab: returns-based strategy decomposition for the 16 shortlist funds 2026-08-26 11:59:59 -04:00
decompose.py Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
edgar_universe.py Find candidate funds NOT in the DB: exhaustive EDGAR 497-universe pass 2026-08-26 15:22:52 -04:00
edgar.py fundlab: extract strategy sections from EDGAR docs; --strategy CLI mode 2026-08-25 23:06:13 -04:00
factor_results.json Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
factors.log Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
factors.py Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
fundinfo.py fundlab: extract strategy sections from EDGAR docs; --strategy CLI mode 2026-08-25 23:06:13 -04:00
funds_curated.json fundlab: EDGAR investment-objective fetcher + curated fallback 2026-08-25 18:15:48 -04:00
harvest.py Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
nport_manifest.json Fund Lab: returns-based strategy decomposition for the 16 shortlist funds 2026-08-26 11:59:59 -04:00
nport.py Fund Lab: N-PORT holdings page for the 16-fund shortlist 2026-08-26 09:59:24 -04:00
overnight.log Overnight comprehensive screen COMPLETE: 2,384 funds, 250 candidates, RESEARCH.md results + next iterations 2026-08-27 06:56:33 -04:00
overnight.py Watchdog: pidfile liveness (pgrep -f was fooled by shells quoting the pattern); per-batch download fault isolation 2026-08-26 22:09:20 -04:00
pool.py fundlab: EDGAR investment-objective fetcher + curated fallback 2026-08-25 18:15:48 -04:00
RESEARCH.md Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
resolved_tickers.json Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
search_all.json Overnight comprehensive screen COMPLETE: 2,384 funds, 250 candidates, RESEARCH.md results + next iterations 2026-08-27 06:56:33 -04:00
search_external.json Find candidate funds NOT in the DB: exhaustive EDGAR 497-universe pass 2026-08-26 15:22:52 -04:00
search_harvest.json Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
search_mined.json Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
search_results.json Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
search.py Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
searchlist.py Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
streamlit.log Factor screen (v2, 35 drivers) + return-driver clusters 2026-08-27 08:35:42 -04:00
tickers.py Alpha search: mine + screen the local DB for idiosyncratic alpha complements 2026-08-26 13:34:34 -04:00
watchdog.log Overnight comprehensive screen COMPLETE: 2,384 funds, 250 candidates, RESEARCH.md results + next iterations 2026-08-27 06:56:33 -04:00
watchdog.sh Watchdog: pidfile liveness (pgrep -f was fooled by shells quoting the pattern); per-batch download fault isolation 2026-08-26 22:09:20 -04:00