本文利用合成控制法 (synthetic control method)，研究網路輿情負面聲量與高端疫苗 (MVC covid-19 vaccine) 施打量之間的因果關係 (causality)。以台灣衛生福利部疾病管制署 (CDC) 提供的疫苗施打期數為基準，在施打區間上做些許的調整，並搭配 OpView 社群口碑資料庫的聲量資料，探討2021年9月24日至10月2日的網路輿情負面聲量介入效果 (treatment effect)，是否對日後的高端疫苗施打量有影響。研究結果顯示，2021年10月3日至10月5日的高端疫苗施打量下降98%，與模擬出來的對照組 (control group) 相比下多減少70%，而10月5日後的高端疫苗施打量也都低於模擬對照組的施打量。

This research studies the causality between the negative public opinion on the internet and the number of MVC covid-19 vaccinated by the synthetic control method (SCM). Based on the vaccine data provided by the Centers for the Disease Control and Prevention (CDC) of the Taiwan Ministry of Health and Welfare, some adjustments are made to the data period in the analysis. Together with the public opinion data of the OpView, this study explores the effect of negative online public opinion on the number of MVC vaccinated from September 24th, 2021 to October 2nd, 2021. The results show a 98% drop in MVC vaccinations between October 3rd and October 5th, 2021, and a 70% reduction compared to the simulated control group. The decrease in the MVC vaccinations has a continuous impact in the later periods.

第一章 緒論 1
第二章 文獻回顧 4
第一節 covid-19對社會的影響 4
第二節 合成控制法的演進 5
第三節 改良後的合成控制法 6
第三章 資料介紹 8
第一節 OpView社群口碑資料庫 - 聲量資料 8
第二節 台灣衛生福利部疾病管制署 - 新冠疫苗資料 9
第三節 資料處理 9
第四節 介入效果 12
第四章 研究方法 16
第五章 實證結果 18
第一節 變數 18
第二節 權重 20
第三節 模擬 21
第四節 安慰劑檢定 22
第六章 結論 24
第一節 研究結果與貢獻 24
第二節 研究限制與未來研究方向 24
參考文獻 26
附錄 I

一、中文文獻：
陳宜廷 (2019), “臺灣與南韓之經濟成長比較-合成控制法下的反事實分析”, 臺灣經濟預測與政策(中央研究院經濟研究所), 50(1), 1-410.


二、英文文獻：
Abadie, A. ＆ J. Gardeazabal (2003), “The Economic Costs of Conflict: A Case Study of the Basque Country.”, The American Economic Review, 93, 112–132.

Abadie, A., A. Diamond & J. Hainmueller (2010), “Synthetic Control Methods for Comparative Case Studies: Estimating the Effect of California’s Tobacco Control Program.”, Journal of the American Statistical Association, 105:490, 493–505.

Abadie, A., A. Diamond & J. Hainmueller (2015), “Comparative Politics and the Synthetic Control Method.”, American Journal of Political Science, 59, 495–510.

Abadie, A. & J. L’Hour (2021), “A Penalized Synthetic Control Estimator for Disaggregated Data.”, Journal of the American Statistical Association, 116:536, 1817-1834.

Abadie, A. (2021), “Using Synthetic Controls: Feasibility, Data Requirements, and Methodological Aspects.”, Journal of Economic Literature, 59(2), 391-425.

Ben-Michael, E., A. Feller & J. Rothstein (2021), “The Augmented Synthetic Control Method.”, Journal of the American Statistical Association, 116:536, 1789-1803.

Chen, Y.-T. (2020), “A distributional synthetic control method for policy evaluation.”, Journal of Applied Econometrics, 35, 505-525.

Chen, Y.-T. (2022), “Regularization of Synthetic Controls for Policy Evaluation.”, Department of Finance National Taiwan University.

Doudchenko, N. & G. W. Imbens (2016), “Balancing, Regression, Difference-in-difference and synthetic control methods: A synthesis.”, NBER Working Paper.

Ferman, B. & C. Pinto (2021), “Synthetic controls with imperfect pretreatment fit.”, Quantitative Economics, 12, 1197-1221.

Fetzer, T., L. Hensel, J. Hermle & C. Roth (2020), “Coronavirus Perception and Economic Anxiety.”, Review of Economics and Statistic, 2021; 103 (5): 968-978.

Saleska, J., L. & Choi, K., R. (2021), “A behavioral economics. perspective on the COVID-19 vaccine amid public mistrust.”, TBM, 11:821-825.

Valero, R. (2015), “Synthetic Control Method versus Standard Statistical Techniques: a Comparison for Labor Market Reforms.”, Working paper, University of Alincante.

Vergura, S. (2020), “Bollinger Bands Based on Exponential Moving Average for Statistical Monitoring of Multi-Array Photovoltaic Systems.”, Energies, 13, 3992.


三、中文書籍
伊藤公一朗(王美娟譯) (2018)，《數據分析的力量》，台灣東販。


四、英文書籍
Cunningham, S. (2021), “Causal inference: The Mixtape”, Yale University.

D.Angrist, J. & Pischke, J.S. (2009), “Most Harmless Econometrics: An Empiricist’s Companion”, Princeton University.