spb/gsf6053 Public

1<!--2=============================================================================3GSF-6053 : Économétrie Financière I4Auteur  : Simon-Pierre Boucher5Contact : contact@spboucher.ai6=============================================================================7-->89# GSF-6053 — Financial Econometrics I (Économétrie Financière)1011**Author / Auteur : Simon-Pierre Boucher**12📧 **Contact : [contact@spboucher.ai](mailto:contact@spboucher.ai)**13🏛️ Département de finance, assurance et immobilier — Faculté des sciences de l'administration, Université Laval1415[![LaTeX](https://img.shields.io/badge/LaTeX-008080?style=for-the-badge&logo=latex&logoColor=white)](https://www.latex-project.org/)16[![Stata](https://img.shields.io/badge/Stata-1F4E79?style=for-the-badge&logo=stata&logoColor=white)](https://www.stata.com/)17[![Course Level](https://img.shields.io/badge/Course%20Level-Graduate-blue)](#)18[![Credits](https://img.shields.io/badge/Credits-3-green)](#)19[![Language](https://img.shields.io/badge/Language-Fran%C3%A7ais-lightgrey)](#)2021---2223## 📚 Course Description2425This repository contains the complete teaching material for **GSF-6053 — Financial Econometrics I**, a graduate course (Hiver 2022) that introduces students to the practical aspects of econometric methods and estimation techniques as applied in finance. All lecture decks are written in **LaTeX (Beamer)**, in French, and each folder ships both the `.tex` source and the compiled `.pdf`.2627The course emphasizes modeling challenges specific to financial applications: OLS regression, maximum likelihood, hypothesis testing, heteroskedasticity, autocorrelation, panel data models, and time-series analysis (ARMA/ARIMA, unit roots, cointegration), with hands-on **Stata** application sessions.2829### 📋 Prerequisites3031A basic understanding of statistics and linear algebra is recommended to engage fully with the course material.3233## 🗂️ Detailed Content by Session3435### Section 02 — Régression et Moindres Carrés Ordinaires (MCO / OLS)3637| Session | Folder | Topics |38|---|---|---|39| Séance 2 | `Séance 2/` | Modèle de régression, dérivation des MCO (sommation et format matriciel), exemples numériques, nuages de points et impact de la variance des erreurs |40| Séance 3 | `Séance 3/` | Maximum de vraisemblance (MLE), MLE pour le modèle de régression linéaire, estimateur sans biais de la variance, propriétés des estimateurs MLE et MCO |41| Séance 4 | `Séance 4/` | Analyse de variance, R², R² ajusté, tests d'hypothèses, contraintes linéaires, tests de Wald, LR (ratio de vraisemblance) et LM (multiplicateur de Lagrange), liens entre F, Wald, LR et LM |42| Séance 5 | `Séance 5/` | Suite de la Section 02 (PDF compilé fourni) |4344### Section 03 — Extensions au modèle linéaire simple4546| Session | Folder | Topics |47|---|---|---|48| Séance 6 | `Séance 6/` | Hétéroscédasticité : définition, conséquences et tests diagnostiques (Breusch-Pagan, White) |49| Séance 7 | `Séance 7/` | Autocorrélation des erreurs : diagnostic (Durbin-Watson), transformations de Cochrane-Orcutt et Prais-Winsten, estimateur robuste de Newey-West |5051### Section 04 — Les modèles panels5253| Session | Folder | Topics |54|---|---|---|55| Séance 8 | `Séance 8/` | Avantages des données panel, notation générale, modèle pooled, modèle à effets fixes, Least Squares Dummy Variable (LSDV) |56| Séance 9 | `Séance 9/` | Modèle empilé, within-group estimator, modèle à effets aléatoires, transformation de Fuller-Battese |5758### Section 05 — Les séries chronologiques5960| Session | Folder | Topics |61|---|---|---|62| Séance 10 | `Séance 10/` | Séries stationnaires et non stationnaires, opérateur de retard, autocovariance et autocorrélation (ACF/PACF), équations de Yule-Walker, modèle AR(1) |63| Séance 11 | `Séance 11/` | Estimation des moments, tests de bruit blanc (Box-Pierce, Ljung-Box), tests de racine unitaire (Dickey-Fuller augmenté, Phillips-Perron), cointégration et tests de cointégration |64| Séance 12 | `Séance 12/` | Processus autorégressifs (AR), moyennes mobiles (MA), ARMA(p,q), ARIMA(p,d,q), méthodologie Box-Jenkins |6566### 💻 Stata Application Sessions6768| Session | Folder | Topics |69|---|---|---|70| STATA S02 | `STATA_S02/` | Application Stata de la Section 02 : statistiques descriptives, régressions MCO, t-test, F-test, analyse des résidus (avec captures d'écran des sorties Stata) |71| STATA S03 | `STATA_S03/` | Application Stata de la Section 03 : détection et correction de l'hétéroscédasticité (tests de Breusch-Pagan et de White, tables du χ²) |7273## 📁 Repository Structure7475```76GSF6053/77├── README.md                  ← this file78├── Séance 2/ … Séance 12/     ← lecture decks (Beamer .tex + compiled .pdf + build files)79│   └── GSF6053_S<n>.tex/.pdf80├── STATA_S02/                 ← Stata lab, Section 02 (deck + Stata output screenshots .png)81│   └── GSF6053_STATA_02.tex/.pdf82└── STATA_S03/                 ← Stata lab, Section 03 (deck + Stata output screenshots .png)83    └── GSF6053_STATA_03.tex/.pdf84```8586Each session folder contains:8788- **`.tex`** — the Beamer source (UTF-8, French)89- **`.pdf`** — the compiled slide deck, ready to use90- **`.png` / `.jpg`** — figures, regression outputs, statistical tables referenced by the deck91- Auxiliary LaTeX build files (`.aux`, `.log`, `.nav`, `.out`, `.snm`, `.toc`, `.synctex.gz`)9293## 🔨 Building the Slides9495All decks compile with a standard TeX distribution (TeX Live / MacTeX). From a session folder:9697```bash98cd "Séance 2"99pdflatex GSF6053_S2.tex100pdflatex GSF6053_S2.tex   # second pass for the table of contents / navigation101```102103Or compile everything at once from the repository root:104105```bash106for d in "Séance "*/ STATA_S0*/; do107  (cd "$d" && f=$(ls *.tex 2>/dev/null) && [ -n "$f" ] && pdflatex -interaction=nonstopmode "$f" && pdflatex -interaction=nonstopmode "$f")108done109```110111**Requirements:** `beamer`, `graphicx`, `inputenc` (utf8) — all included in any full TeX Live / MacTeX install. Images must stay next to their `.tex` file (they are referenced by relative path).112113## 🎓 Learning Outcomes114115By the end of this course, students will be able to:116117- Derive and apply the OLS and maximum-likelihood estimators to financial data;118- Perform and interpret hypothesis tests (t, F, Wald, LR, LM);119- Diagnose and correct heteroskedasticity and autocorrelation;120- Estimate panel-data models (pooled, fixed effects, random effects);121- Model financial time series (AR, MA, ARMA, ARIMA), test for unit roots and cointegration, and apply the Box-Jenkins methodology;122- Implement all of the above in Stata.123124## 🤝 Contributing125126This repository is primarily for educational purposes. If you find any errors or have suggestions for improvements, feel free to open an issue or submit a pull request.127128## 📄 License & Usage129130This material is intended for students enrolled in the GSF-6053 course and for educational use. For any other use, please contact the author.131132---133134**Simon-Pierre Boucher** — 📧 [contact@spboucher.ai](mailto:contact@spboucher.ai)135136⭐ *Star this repository if you find it helpful!* ⭐137