Phase 1: foundation — simulators, data layer, chain scraper - #1
Merged
Conversation
Python 3.12 via uv with committed lockfile. All numeric assumptions live in configs/default.yaml; CI gates on lint, format, strict types and 85% coverage.
…ture noise Exact log-normal GBM stepping; Heston via full-truncation Euler (Lord, Koekkoek & van Dijk 2010) recording per-path integrated variance, cross-checked against the analytic CIR mean; Merton (1976) with compensated drift and per-jump bookkeeping so jump variation is known exactly; additive noise (Zhang-Mykland-Ait-Sahalia 2005 setting) and Roll (1984) bid-ask bounce. Self-consistency suite: moment matching within Monte Carlo error, martingale checks, sum-of-squared-returns vs integrated variance, the naive-RV-diverges-under-noise signature, and Roll negative first-order autocovariance. Property tests via hypothesis.
…uality metrics Dukascopy .bi5 decoded directly in Python (ADR 0003): resumable per-day parquet with retries and throttling; decoder unit-tested against crafted payloads, empty weekend hours are valid. FRED fredgraph parser keeps missing markers as nulls; ECB SDMX parser for the EONIA->ESTR splice. CBOE delayed-quotes scraper for FXE/_SPX and minor currency ETFs (ADR 0002), OCC symbology parsed from the fixed-width tail, quote timestamps stored ET->UTC. Chain cleaning logs every dropped row by reason and reconciles input = output + drops as a unit test. Per-day tick quality metrics feed the report script.
…nt per-day ingestion
…ality report The 2020-03-16..25 hour files (peak COVID tick volume) exceed a 30s request timeout; 90s ingests the whole sample window with zero failures. Quality report covers 26 days of March 2020 plus the first five chain snapshots.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What was built
Test results
56 passed, 0 failed; coverage 90.75% (gate 85%); mypy --strict clean on 17 source files; ruff clean. Ground-truth self-consistency includes: GBM/Merton moment and martingale checks against Monte Carlo error bounds, Heston integrated variance vs the analytic CIR mean, the naive-RV-diverges-under-noise signature (demonstrated, not assumed), Roll negative first-order autocovariance, .bi5 decode round-trips, and chain-cleaning reconciliation (input = output + logged drops).
Known limitations / uncertainties