.zenodo.json

{
  "title": "Modelling Volatility and Risk Spillover Between the Financial Markets of US and China Using GARCH Value-at-Risk Forecasting and Granger Causality",
  "upload_type": "publication",
  "publication_type": "thesis",
  "publication_date": "2020-06-01",
  "version": "Editorially revised 2026",
  "language": "eng",
  "access_right": "open",
  "license": "cc-by-4.0",
  "thesis_university": "Seoul National University, Department of Economics",
  "creators": [
    {
      "name": "Suh, Jongsun",
      "affiliation": "Seoul National University",
      "orcid": "0009-0003-5053-4246"
    }
  ],
  "keywords": [
    "GARCH",
    "ARIMA",
    "value-at-risk",
    "Granger causality",
    "volatility forecasting",
    "risk spillover",
    "financial econometrics",
    "US-China markets"
  ],
  "description": "<p>Volatility forecasting and risk spillover evaluation is performed on a dataset consisting of the intraday returns from January 2007 to April 2020 of the S&amp;P500, SSE Composite Index, and the Chinese Yuan to USD exchange rate. The chosen date range includes both the Great Recession of 2008 and the more recent Coronavirus Recession, offering a rare opportunity for comparative analysis of those two periods of elevated volatility against the overall backdrop of a strong, sustained bull market. For volatility forecasting, a Skewed Student's t ARIMA-GARCH model is used to estimate and forecast Value-at-Risk series, which are in turn used to test for the presence of Granger Causality between the US and Chinese financial markets. The GARCH-VaR method exhibits a high fit to historical data, and while a considerable degree of risk spillover is observed between the US and Chinese economies throughout the date range, its predictive power is shown to markedly diminish during the two Recession periods.</p><p><em>Undergraduate thesis, Department of Economics, Seoul National University, completed June 2020. Editorially revised 2026: byline, numeric presentation, and prose only. Results, methods, figures, and findings are unchanged from the submitted version. This deposit archives the thesis together with the analysis notebooks that produced it.</em></p>"
}