TeX 62.4%
Python 37.5%
1<!-- Author: Simon-Pierre Boucher — simon-pierre.boucher@uqo.ca -->2# 🌐 Intraday Volatility Transmission Across Asset Classes34**UQO Working Paper No. 6** — *A Connectedness Analysis of Equities, Fixed Income, Commodities, and Cryptocurrencies*56[-b31b1b?logo=latex&logoColor=white)](paper/main.pdf)7[](requirements.txt)8[](data/raw/README.md)9[](scripts/)10[](https://uqo.ca)11[](mailto:simon-pierre.boucher@uqo.ca)1213> **TL;DR** — Daily realized volatility is built from 5-minute bars (2008–2026) for 10 instruments spanning equity futures (ES, NQ), FX (EURUSD, USDJPY, GBPUSD), commodities (GC, CL), crypto (BTC, ETH) and the VIX, and fed to the Diebold–Yilmaz (2012) connectedness framework (VAR(4), 10-day GFEVD). Total spillover is **55.4 %**; ES and NQ are the dominant transmitters (**+26.8 / +19.7 pp** net), the VIX the main receiver (**−23.4 pp**), and crypto stays largely isolated (own-shock share > 77 %). Connectedness peaks at **81.6 %** in February 2018; during COVID USD/JPY becomes the top shock originator and in the 2022 hiking cycle Bitcoin turns into a net transmitter (+7.9 pp).1415---1617## 📌 Status1819| Item | Value |20|---|---|21| Version | 1.0 (May 2026) |22| SSRN | not yet submitted |23| Journal | — |24| Code & data | reproducible from `scripts/01_analysis.py`; raw DuckDB files not redistributed |2526## 🎯 What this paper does2728The paper measures how volatility shocks travel across asset classes at the daily frequency, using realized volatility computed from intraday data rather than low-frequency proxies. Ten representative instruments are followed over 2008–2026 and the generalized forecast-error variance decomposition of a VAR is used to build directional, net and total spillover indices:2930```31RV_{i,t} = Σ_k r²_{i,t,k} (5-min returns) → VAR(4) on log RV → GFEVD(H = 10) → TSI, TO, FROM, NET32```3334Full-sample, rolling (200-day) and regime-specific (pre-COVID, COVID crash, 2022 rate hikes) decompositions are complemented by a trading-session analysis (Asian / European / U.S. hours) and Granger-causality tests between equity futures, the VIX and crypto.3536## 🏆 Headline results3738| Object | Estimate |39|---|---|40| Total spillover index, full sample | 55.44 % |41| Net spillover ES / NQ | +26.80 / +19.72 pp |42| Net spillover VIX | −23.37 pp |43| Own-shock share BTC, ETH | > 77 % |44| Rolling TSI peak | 81.55 % (Feb 2018 volatility implosion) |45| COVID crash top transmitter | USD/JPY (+60.29 pp) |46| 2022 rate hikes: BTC net spillover | +7.90 pp (net transmitter) |47| Granger: equity futures → VIX | F = 19.86, p < 0.001 |4849## 🗂 Repository layout5051```52intraday-volatility-connectedness/53├── data/54│ ├── raw/ # futures_5min / fx_5min / crypto_5min .duckdb (see data/raw/README.md) — not committed55│ └── processed/56├── scripts/57│ └── 01_analysis.py # RV construction, VAR/GFEVD, rolling & regime analysis, sessions, Granger, figures58├── results/ # table1…table4 CSV, daily_rv_all.csv, rolling_total_spillover.csv59├── figures/ # fig1…fig7 PNG (copied into paper/figures)60└── paper/ # main.tex + sections/ + appendix/ → main.pdf61```6263## 🚀 Quick start6465```bash66python3 -m pip install -r requirements.txt67# place the three DuckDB files in data/raw/ (or export WP6_DATA_DIR=/path/to/dbs)68python3 scripts/01_analysis.py # ≈ 10 min, ≈ 8 GB RAM (loads 235 M bars)69cd paper && make # latexmk → main.pdf70```7172## 🗃 Data73745-minute OHLCV bars from FirstRate Data, 2008–2026: CME/CBOE continuous futures (ES, NQ, GC, CL, VX), spot FX (EURUSD, USDJPY, GBPUSD) and crypto spot (BTC, ETH). Realized volatility is the daily sum of squared 5-minute log returns; sessions are defined in exchange time. See `data/raw/README.md`.7576## 🔬 Methodology77781. **Realized volatility** per instrument and day from 5-minute closes (`scripts/01_analysis.py`, §1–2).792. **Diebold–Yilmaz (2012) connectedness**: VAR(4) on log RV, generalized FEVD at H = 10, total / directional / net spillovers (§3).803. **Rolling 200-day** windows for time variation (§4).814. **Regime sub-samples**: pre-COVID, COVID crash, 2022 rate hikes (§5).825. **Trading sessions**: hourly volatility profiles and session-level Granger causality (§6).836. **Network graph** of net pairwise spillovers (§7).8485## 🖼 Figures & tables inventory8687| Figure | Content |88|---|---|89| `fig1_rv_timeseries` | daily realized volatility, 10 instruments |90| `fig2_connectedness_heatmap` | full-sample spillover table |91| `fig3_rolling_spillover` | rolling total spillover index |92| `fig4_net_spillovers_regimes` | net spillovers by regime |93| `fig5_hourly_volatility` | intraday volatility by session |94| `fig6_net_directional_btc_es` | net directional spillovers BTC ↔ ES |95| `fig7_network_graph` | net pairwise spillover network |9697Tables (`results/*.csv`): descriptive statistics, full connectedness table, three regime tables, session Granger tests.9899## ⚠️ Limitations100101- Continuous futures series splice contracts at roll dates; roll returns are not removed.102- Crypto trades 24/7 while futures and FX have session gaps, so daily RV windows differ across classes.103- The GFEVD depends on the VAR lag (4) and horizon (10); robustness to these choices is reported in the appendix only.104105## 📝 Citation106107```bibtex108@techreport{boucher2026connectedness,109 author = {Boucher, Simon-Pierre},110 title = {Intraday Volatility Transmission Across Asset Classes: A Connectedness Analysis of Equities, Fixed Income, Commodities, and Cryptocurrencies},111 institution = {Universit\'e du Qu\'ebec en Outaouais, D\'epartement des sciences administratives},112 type = {Working Paper},113 number = {6},114 year = {2026},115 month = {May}116}117```118119## 👤 Author & contact120121**Simon-Pierre Boucher** — Département des sciences administratives,122Université du Québec en Outaouais (UQO), Gatineau, QC.123📧 [simon-pierre.boucher@uqo.ca](mailto:simon-pierre.boucher@uqo.ca)124125## 🗒 Notes126127- References are typeset with an inline `thebibliography` (`paper/sections/references.tex`, apalike style), not a `.bib` file — kept as is (deviation from `HARMONISATION.md` §3).128- The original flat manuscript (`paper.tex`, May 2026) is archived in `Recherche-UQO/_archives/versions-mai-2026/intraday-volatility-connectedness--flat-tex-20260520/`.129