https://viethungvu1998.github.io/Recent content on Viet Hung VuHugoen-usWed, 03 Jun 2026 09:00:00 +1000https://viethungvu1998.github.io/posts/quantitative-finance-notebook/Wed, 03 Jun 2026 09:00:00 +1000https://viethungvu1998.github.io/posts/quantitative-finance-notebook/<p>I want this site to become a working notebook for quantitative finance rather than a collection of polished claims.</p> <p>The main goal is to make ideas testable. When I write about a signal, a backtest, or a risk model, I want to be clear about the data assumptions, evaluation method, and limitations.</p>https://viethungvu1998.github.io/posts/backtesting-checklist/Tue, 02 Jun 2026 09:00:00 +1000https://viethungvu1998.github.io/posts/backtesting-checklist/<p>Backtesting is useful only when the experiment is honest. Before trusting a result, I want to check the assumptions that can quietly inflate performance.</p> <p>The basic checklist: define the universe before the test, avoid look-ahead bias, model transaction costs, handle missing data carefully, separate training and evaluation periods, and report drawdowns alongside returns.</p>https://viethungvu1998.github.io/posts/factor-model-study-plan/Mon, 01 Jun 2026 09:00:00 +1000https://viethungvu1998.github.io/posts/factor-model-study-plan/<p>Factor models are a useful starting point for thinking about risk, return, and portfolio construction.</p> <p>My study plan is to begin with market beta, move to multi-factor models, then focus on how factors are estimated, tested, combined, and monitored through time.</p>https://viethungvu1998.github.io/posts/volatility-risk-notes/Sun, 31 May 2026 09:00:00 +1000https://viethungvu1998.github.io/posts/volatility-risk-notes/<p>Volatility is both a modeling target and a practical risk signal.</p> <p>In a research workflow, I care less about finding a single perfect volatility model and more about comparing forecasts under consistent assumptions. Realized volatility, historical volatility, GARCH-style models, and machine learning forecasts should be evaluated on the same data splits and loss functions.</p>https://viethungvu1998.github.io/posts/market-data-cleaning/Sat, 30 May 2026 09:00:00 +1000https://viethungvu1998.github.io/posts/market-data-cleaning/<p>Market data cleaning is not a boring preprocessing step. It is part of the research problem.</p> <p>Bad timestamps, missing prices, duplicated rows, split adjustments, survivorship bias, and inconsistent symbol histories can all change the conclusion of a strategy test.</p>https://viethungvu1998.github.io/about/Mon, 01 Jan 0001 00:00:00 +0000https://viethungvu1998.github.io/about/<p>I am Viet Hung Vu, a PhD candidate at Griffith University, Australia. My background is in computer science, data science, and AI research, and I use this site to develop a public notebook around quantitative finance.</p>