查詢結果分析
來源資料
相關文獻
- 電動車龍頭的市場韌性:應用GJR-GARCH模型探討特斯拉股價波動性與外生因子之關聯性
- Using Three Value-at-Risk Models to Measure the Financial Risk of Various Portfolios Including Taiwan and Hong Kong Stock Indices and International Currencies during the Asian Crisis Period
- 報酬與波動性動態關聯之研究--摩根臺股指數與指數期貨之探討
- 監視制度對平均數復歸與報酬波動性影響之研究
- 臺股指數現貨與期貨日內報酬波動不對稱關聯性之研究
- 監視制度對雜訊交易與報酬波動性影響之實證研究
- The Impact of Trading Volume on Stock Returns and Volatility:An Examination of Taiwan and Hong Kong Data
- The Impact of Trading Volume on Stock Returns and Volatility: An Examination of Taiwan and Hong Kong Data
- 投資型壽險商品購買行為之研究--消費者風險價值觀實證
- 金磚五國之期貨避險績效--動態Copula-GJR-GARCH模型應用
頁籤選單縮合
| 題 名 | 電動車龍頭的市場韌性:應用GJR-GARCH模型探討特斯拉股價波動性與外生因子之關聯性=Market Resilience of an Electric Vehicle Leader: Investigating the Relationship between Tesla's Stock Price Volatility and Exogenous Factors Using the GJR-GARCH Model |
|---|---|
| 作 者 | 蔡裕明; 林凱馨; 哈宜蓁; | 書刊名 | 管理資訊計算 |
| 卷 期 | 14:2 2025.09[民114.09] |
| 頁 次 | 頁93-108 |
| 分類號 | 447.21 |
| 關鍵詞 | 特斯拉; GJR-GARCH模型; 報酬波動性; 風險價值; 產業競爭與市場因子; Tesla; GJR-GARCH model; Return volatility; Value at risk; VaR; Industry competition; Market factors; |
| 語 文 | 中文(Chinese) |
| DOI | 10.6285/MIC.202509_14(2).0009 |
| 中文摘要 | 本研究旨在探討影響特斯拉(Tesla)股價報酬與波動性之主要市場與產業外部因素,並透過GARCH家族模型建構波動性與風險預測模型。研究樣本涵蓋2021至2025年3月之資料,納入S&P500報酬、VIX指數、10年期美債殖利率、主要競爭對手股價(如NIO、RIVN)以及鋰電池原料價格(如碳酸鋰)作為外生解釋變數。初步以OLS模型檢驗報酬線性關係,結果顯示整體市場走勢與競爭對手報酬對特斯拉股價具顯著正向影響。接著,本文採用GARCH、GARCH-X、EGARCH、GJR-GARCH與GJR-GARCH-X模型進行比較,並以AIC、BIC與風險價值(Value at Risk, VaR)回溯測試(包含Kupiec與Christoffersen統計檢定)作為模型績效之綜合評估依據。實證結果顯示,GJR-GARCH(1,1)模型在模型配適與風險預測方面表現最佳,具備最低之AIC與BIC值,並通過VaR違約比率檢定,顯示其能同時捕捉報酬序列之非對稱性(槓桿效應)與外部市場因子對波動結構性影響,具有高度統計穩健性與經濟直觀性。此外,標準化殘差診斷亦顯示特斯拉報酬序列存在顯著之波動性聚集與負向非對稱反應,說明其對壞消息具有較高敏感性。整體而言,本研究所建構之GJR-GARCH模型,具備良好之解釋力與預測能力,對於科技成長股之波動建模、風險評估、資產配置策略與動態風控具實務應用價值。未來可進一步納入社群情緒變數或延伸至多變量GARCH架構,俾提升模型深度與預測效能。 |
| 英文摘要 | This study investigates the key market and industry-level external factors affecting Tesla’s stock returns and volatility by constructing a risk forecasting framework using GARCH-family models. The sample covers data from 2021 to March 2025, incorporating S&P 500 returns, the VIX index, 10-year U.S. Treasury yields, competitor stock prices (e.g., NIO, RIVN), and lithium battery raw material prices (e.g., lithium carbonate) as exogenous variables. An initial OLS analysis reveals that both overall market trends and competitor performance have significant positive impacts on Tesla’s returns. Subsequently, five volatility models—GARCH, GARCH-X, EGARCH, GJR-GARCH, and GJR-GARCHX—are compared using AIC, BIC, and Value at Risk (VaR) backtesting, including Kupiec and Christoffersen tests. The results indicate that the GJR-GARCH(1,1) model outperforms others in both model fit and risk prediction, effectively capturing return asymmetry and external market influence on volatility. Diagnostic analysis also confirms the presence of volatility clustering and negative asymmetry in Tesla’s returns, indicating heightened sensitivity to negative news. This model demonstrates strong explanatory and predictive power, offering practical value for volatility modeling, risk assessment, and dynamic asset management. Future research may incorporate sentiment variables or extend to multivariate GARCH frameworks. |
本系統中英文摘要資訊取自各篇刊載內容。