學術產出-Theses
Article View/Open
Publication Export
-
題名 臺灣產業類股間因果關係之研究：以雙重變數選擇過程檢驗高維度 Granger 因果關係
Causal Relationship between Sector Indices in Taiwan Stock Market：Testing High-Dimensional Granger Causality with a Post-Double-Selection Procedure作者 劉芳均
Liou, Fang-Jun貢獻者 徐士勛
Hsu, Shih-Hsun
劉芳均
Liou, Fang-Jun關鍵詞 產業類股報酬關係
高維度Granger因果關係
LASSO
雙重變數選擇過程
邊緣介數法
Sector Indices Relationship
HD-Granger Causality
LASSO
Post Double Selection Procedure
Edge Betweenness日期 2023 上傳時間 2-Aug-2023 13:41:26 (UTC+8) 摘要 本研究採用 COVID-19 疫情期間的資料探討 27 個台股產業指數週報酬間的因果關係，並分別探討美國聯準會實施量化寬鬆和量化緊縮政策兩段期間資金面的變化對產業股價報酬關係的影響。本文主要根據 Hecq et al. (2019) 提出的分析架構，透過 LASSO 方法進行雙重變數選擇對原高維度 VAR 模型進行系統降維，以得到更為穩健的估計與推論，最後再利用網路圖呈現 Granger 因果關係和分群結果。首先，在量化寬鬆期間的分析中，我們發現受惠於疫情的生技醫療、食品和資訊服務業類股的股價報酬領先於其他產業，且估計係數幾乎為正，顯示這三個產業在當時為潛在領漲的類股。而電器電纜、半導體、光電類股的股價報酬亦領先於其他產業，但估計係數幾乎為負，多數電子類股在此期間的負向領先關係顯示了在此期間為潛在領跌族群。相對地，在量化緊縮期間，我們發現建材營造和金融保險類股為相對領先者，並且皆對其他產業的估計係數正負相間，顯示了各類股表現不一，跟量化寬鬆期間相比，此期間可能受到通膨和升息等較雜亂的市場訊息影響。綜合分析上述兩段期間的分群結果，在顯著水準為 1% 之下，均將網路分成多群大小相異的族群；在顯著水準為 5% 之下，均將較關聯的產業分成一個大族群，以及其餘的獨立族群，兩種類型的結果分別適合關注小群類股以及整體產業趨勢的研究者。
This research aims to analyze the causal relationship between the weekly returns of 27 stock sector indices in Taiwan stock market during the COVID-19 epidemic period by testing high-dimensional Granger causality with a post-double-selection procedure. The analysis framework is primarily based on the methodology proposed by Hecq et al. (2019), and the findings are presented through network graphs.Our findings indicate that during the period of quantitative easing, the returns of biotechnology and medical, food and information services sectors primarily Granger caused the returns of other industries, suggesting that these sectors were potential leading stocks at that time. On the other hand, during the period of quantitative tightening, our findings indicate that the returns of construction and financial sectors primarily Granger caused the returns of other industries. However, the estimated coefficients showed positive or negative values, indicating varied performances across sectors.Fianaly, the clustering results indicate under significance levels of 1% and 5%, respectively, the former consistently partitioned the network into multiple communities of varying sizes. In contrast, the latter resulted in a single large community and the remaining independent communities.參考文獻 徐士勛與李佳磬 (2019)，「亞洲主要股市報酬波動的潛在鏈結程度衡量」，《經濟論文叢刊》，47(4)， 579-620。郭維裕，李淯靖，陳致綱與林建秀 (2015)，「臺灣產業指數的外溢效果」，《經濟論文叢刊》，43(4)， 407-442。黃裕烈與管中閔 (2014)，《向量自我迴歸模型，計量方法與 R 程式》，台北: 雙葉書廊。鄒易憑與白東岳 (2009)，「原油價格與原油產業指數之動態關係: 厚尾跳躍模型之應用」，《臺灣金融財務季刊》，10(3)， 87-111。蔡坤旻 (2009)，《原油價格變動對於太陽能產業指數的影響-雙門檻 GARCH 模型之應用》，臺北大學國際企業研究所學位論文。Belloni, A., Chernozhukov, V., and Hansen, C. (2014), “High-dimensional methods and inference on structural and treatment effects,” Journal of Economic Perspectives, 28(2), 29-50.Belloni, A., Chernozhukov, V., and Hansen, C. (2014), “Inference on treatment effects after selection among high dimensional controls,” The Review of Economic Studies, 81(2), 608-650.Cao, D., Long, W., and Yang, W. (2014), “Sector indices correlation analysis in China’s stock market,” Procedia Computer Science, 17, 1241-1249.Cavaliere, G., D. I. Harvey, S. J. Leybourne, and A. R. Taylor (2011), “Testing for unit roots in the presence of a possible break in trend and nonstationary volatility,” Econometric Theory, 27(5), 957-991.Chen, Y., Li, W., and Qu, F. (2019), “Dynamic asymmetric spillovers and volatility interdependence on China’s stock market,” Physica A: Statistical Mechanics and its Applications, 523, 825-838.Corsi, F., Lillo, F., Pirino, D., and Trapin, L. (2018), “Measuring the propagation of financial distress with Granger-causality tail risk networks,” Journal of Financial Stability, 38, 18-36.Granger, C. W. (1980), “Testing for causality: A personal viewpoint,” Journal of Economic Dynamics and control, 2, 329-352.Hao, J., and He, F. (2018), “Univariate dependence among sectors in Chinese stock market and systemic risk implication,” Physica A: Statistical Mechanics and its Applications, 510, 355-364.Hecq, A., Margaritella, L., and Smeekes, S. (2019), “Granger causality testing in high-dimensional VARs: a post-double-selection procedure,” arXiv preprint arXiv:1902.10991.Hecq, A., Margaritella, L., and Smeekes, S. (2023), “Inference in Non-stationary High-Dimensional VARs,” arXiv preprint arXiv:2302.01434.Laopodis, N. T. (2016), “Industry returns, market returns and economic fundamentals: Evidence for the United States,” Economic Modelling, 53, 89-106.Leeb, H., and Pötscher, B. M. (2005), “Model selection and inference: Facts and fiction,” Econometric Theory, 21(1), 21-59.Newman, M. E., and Girvan, M. (2004), “Finding and evaluating community structure in networks,” Physical review E, 69(2), 026113.Shahzad, S. J. H., Naeem, M. A., Peng, Z., and Bouri, E. (2022), “Asymmetric volatility spillover among Chinese sectors during COVID-19,” International Review of Financial Analysis, 75, 101754.Shirokikh, O., Pastukhov, G., Semenov, A., Butenko, S., Veremyev, A., Pasiliao, E. L., and Boginski, V. (2022), “Networks of causal relationships in the US stock market,” Dependence Modeling, 10(1), 177-190.Song, X., and Taamouti, A. (2019), “A better understanding of granger causality analysis: A big data environment,” Oxford Bulletin of Economics and Statistics, 81(4), 911-936.Tibshirani, R. (1996), “Regression shrinkage and selection via the lasso” Journal of the Royal Statistical Society: Series B (Methodological), 58(1), 267-288.Toda, H. Y., and Yamamoto, T. (1995), “Statistical inference in vector autoregressions with possibly integrated processes,” Journal of econometrics, 66(1-2), 225-250.Výrost, T., Lyócsa, Š., and Baumöhl, E. (2015), “Granger causality stock market networks: Temporal proximity and preferential attachment” Physica A: Statistical Mechanics and its Applications, 427, 262-276.Wilms, I., S. Gelper, and C. Croux. (2016), “The predictive power of the business and bank sentiment of firms: A high-dimensional Granger Causality approach” European Journal of Operational Research, 254(1), 138-147.Zhou, X., Zhang, H., Zheng, S., Xing, W., Yang, H., and Zhao, Y. (2023), “A study on the transmission of trade behavior of global nickel products from the perspective of the industrial chain,” Resources Policy, 81, 103376. 描述 碩士
國立政治大學
經濟學系
110258004資料來源 http://thesis.lib.nccu.edu.tw/record/#G0110258004 資料類型 thesis dc.contributor.advisor 徐士勛 zh_TW dc.contributor.advisor Hsu, Shih-Hsun en_US dc.contributor.author (Authors) 劉芳均 zh_TW dc.contributor.author (Authors) Liou, Fang-Jun en_US dc.creator (作者) 劉芳均 zh_TW dc.creator (作者) Liou, Fang-Jun en_US dc.date (日期) 2023 en_US dc.date.accessioned 2-Aug-2023 13:41:26 (UTC+8) - dc.date.available 2-Aug-2023 13:41:26 (UTC+8) - dc.date.issued (上傳時間) 2-Aug-2023 13:41:26 (UTC+8) - dc.identifier (Other Identifiers) G0110258004 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/146471 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 經濟學系 zh_TW dc.description (描述) 110258004 zh_TW dc.description.abstract (摘要) 本研究採用 COVID-19 疫情期間的資料探討 27 個台股產業指數週報酬間的因果關係，並分別探討美國聯準會實施量化寬鬆和量化緊縮政策兩段期間資金面的變化對產業股價報酬關係的影響。本文主要根據 Hecq et al. (2019) 提出的分析架構，透過 LASSO 方法進行雙重變數選擇對原高維度 VAR 模型進行系統降維，以得到更為穩健的估計與推論，最後再利用網路圖呈現 Granger 因果關係和分群結果。首先，在量化寬鬆期間的分析中，我們發現受惠於疫情的生技醫療、食品和資訊服務業類股的股價報酬領先於其他產業，且估計係數幾乎為正，顯示這三個產業在當時為潛在領漲的類股。而電器電纜、半導體、光電類股的股價報酬亦領先於其他產業，但估計係數幾乎為負，多數電子類股在此期間的負向領先關係顯示了在此期間為潛在領跌族群。相對地，在量化緊縮期間，我們發現建材營造和金融保險類股為相對領先者，並且皆對其他產業的估計係數正負相間，顯示了各類股表現不一，跟量化寬鬆期間相比，此期間可能受到通膨和升息等較雜亂的市場訊息影響。綜合分析上述兩段期間的分群結果，在顯著水準為 1% 之下，均將網路分成多群大小相異的族群；在顯著水準為 5% 之下，均將較關聯的產業分成一個大族群，以及其餘的獨立族群，兩種類型的結果分別適合關注小群類股以及整體產業趨勢的研究者。 zh_TW dc.description.abstract (摘要) This research aims to analyze the causal relationship between the weekly returns of 27 stock sector indices in Taiwan stock market during the COVID-19 epidemic period by testing high-dimensional Granger causality with a post-double-selection procedure. The analysis framework is primarily based on the methodology proposed by Hecq et al. (2019), and the findings are presented through network graphs.Our findings indicate that during the period of quantitative easing, the returns of biotechnology and medical, food and information services sectors primarily Granger caused the returns of other industries, suggesting that these sectors were potential leading stocks at that time. On the other hand, during the period of quantitative tightening, our findings indicate that the returns of construction and financial sectors primarily Granger caused the returns of other industries. However, the estimated coefficients showed positive or negative values, indicating varied performances across sectors.Fianaly, the clustering results indicate under significance levels of 1% and 5%, respectively, the former consistently partitioned the network into multiple communities of varying sizes. In contrast, the latter resulted in a single large community and the remaining independent communities. en_US dc.description.tableofcontents 1 緒論 11.1 研究背景與目的 11.2 研究架構 32 文獻回顧53 實證模型103.1 高維度 Granger 因果關係檢定 103.2 LASSO 43.3 雙重變數選擇後推論 173.4 非定態高維度 Granger 因果關係檢定 223.5 有向網路圖和鄰近矩陣 243.6 邊緣介數方法 264 實證資料 284.1 資料說明 284.2 資料分析 305 實證結果 355.1 量化寬鬆期間分析結果 365.2 量化緊縮期間分析結果 416 結論與建議 46參考文獻 50附錄 54A 非定態 Granger 因果關係 54B 臺灣產業指數一覽表與資料分析結果 56C 臺灣產業指數相關係數熱力圖結果 64D 未控制美股 Granger 因果關係網路圖結果 66E 疫情前後 Granger 因果關係網路圖結果 69 zh_TW dc.format.extent 1913908 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0110258004 en_US dc.subject (關鍵詞) 產業類股報酬關係 zh_TW dc.subject (關鍵詞) 高維度Granger因果關係 zh_TW dc.subject (關鍵詞) LASSO zh_TW dc.subject (關鍵詞) 雙重變數選擇過程 zh_TW dc.subject (關鍵詞) 邊緣介數法 zh_TW dc.subject (關鍵詞) Sector Indices Relationship en_US dc.subject (關鍵詞) HD-Granger Causality en_US dc.subject (關鍵詞) LASSO en_US dc.subject (關鍵詞) Post Double Selection Procedure en_US dc.subject (關鍵詞) Edge Betweenness en_US dc.title (題名) 臺灣產業類股間因果關係之研究：以雙重變數選擇過程檢驗高維度 Granger 因果關係 zh_TW dc.title (題名) Causal Relationship between Sector Indices in Taiwan Stock Market：Testing High-Dimensional Granger Causality with a Post-Double-Selection Procedure en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) 徐士勛與李佳磬 (2019)，「亞洲主要股市報酬波動的潛在鏈結程度衡量」，《經濟論文叢刊》，47(4)， 579-620。郭維裕，李淯靖，陳致綱與林建秀 (2015)，「臺灣產業指數的外溢效果」，《經濟論文叢刊》，43(4)， 407-442。黃裕烈與管中閔 (2014)，《向量自我迴歸模型，計量方法與 R 程式》，台北: 雙葉書廊。鄒易憑與白東岳 (2009)，「原油價格與原油產業指數之動態關係: 厚尾跳躍模型之應用」，《臺灣金融財務季刊》，10(3)， 87-111。蔡坤旻 (2009)，《原油價格變動對於太陽能產業指數的影響-雙門檻 GARCH 模型之應用》，臺北大學國際企業研究所學位論文。Belloni, A., Chernozhukov, V., and Hansen, C. (2014), “High-dimensional methods and inference on structural and treatment effects,” Journal of Economic Perspectives, 28(2), 29-50.Belloni, A., Chernozhukov, V., and Hansen, C. (2014), “Inference on treatment effects after selection among high dimensional controls,” The Review of Economic Studies, 81(2), 608-650.Cao, D., Long, W., and Yang, W. (2014), “Sector indices correlation analysis in China’s stock market,” Procedia Computer Science, 17, 1241-1249.Cavaliere, G., D. I. Harvey, S. J. Leybourne, and A. R. Taylor (2011), “Testing for unit roots in the presence of a possible break in trend and nonstationary volatility,” Econometric Theory, 27(5), 957-991.Chen, Y., Li, W., and Qu, F. (2019), “Dynamic asymmetric spillovers and volatility interdependence on China’s stock market,” Physica A: Statistical Mechanics and its Applications, 523, 825-838.Corsi, F., Lillo, F., Pirino, D., and Trapin, L. (2018), “Measuring the propagation of financial distress with Granger-causality tail risk networks,” Journal of Financial Stability, 38, 18-36.Granger, C. W. (1980), “Testing for causality: A personal viewpoint,” Journal of Economic Dynamics and control, 2, 329-352.Hao, J., and He, F. (2018), “Univariate dependence among sectors in Chinese stock market and systemic risk implication,” Physica A: Statistical Mechanics and its Applications, 510, 355-364.Hecq, A., Margaritella, L., and Smeekes, S. (2019), “Granger causality testing in high-dimensional VARs: a post-double-selection procedure,” arXiv preprint arXiv:1902.10991.Hecq, A., Margaritella, L., and Smeekes, S. (2023), “Inference in Non-stationary High-Dimensional VARs,” arXiv preprint arXiv:2302.01434.Laopodis, N. T. (2016), “Industry returns, market returns and economic fundamentals: Evidence for the United States,” Economic Modelling, 53, 89-106.Leeb, H., and Pötscher, B. M. (2005), “Model selection and inference: Facts and fiction,” Econometric Theory, 21(1), 21-59.Newman, M. E., and Girvan, M. (2004), “Finding and evaluating community structure in networks,” Physical review E, 69(2), 026113.Shahzad, S. J. H., Naeem, M. A., Peng, Z., and Bouri, E. (2022), “Asymmetric volatility spillover among Chinese sectors during COVID-19,” International Review of Financial Analysis, 75, 101754.Shirokikh, O., Pastukhov, G., Semenov, A., Butenko, S., Veremyev, A., Pasiliao, E. L., and Boginski, V. (2022), “Networks of causal relationships in the US stock market,” Dependence Modeling, 10(1), 177-190.Song, X., and Taamouti, A. (2019), “A better understanding of granger causality analysis: A big data environment,” Oxford Bulletin of Economics and Statistics, 81(4), 911-936.Tibshirani, R. (1996), “Regression shrinkage and selection via the lasso” Journal of the Royal Statistical Society: Series B (Methodological), 58(1), 267-288.Toda, H. Y., and Yamamoto, T. (1995), “Statistical inference in vector autoregressions with possibly integrated processes,” Journal of econometrics, 66(1-2), 225-250.Výrost, T., Lyócsa, Š., and Baumöhl, E. (2015), “Granger causality stock market networks: Temporal proximity and preferential attachment” Physica A: Statistical Mechanics and its Applications, 427, 262-276.Wilms, I., S. Gelper, and C. Croux. (2016), “The predictive power of the business and bank sentiment of firms: A high-dimensional Granger Causality approach” European Journal of Operational Research, 254(1), 138-147.Zhou, X., Zhang, H., Zheng, S., Xing, W., Yang, H., and Zhao, Y. (2023), “A study on the transmission of trade behavior of global nickel products from the perspective of the industrial chain,” Resources Policy, 81, 103376. zh_TW
