WEKO3
アイテム
APPLYING MACHINE LEARNING TO UNCOVER PATTERNS IN ECONOMIC ISSUES: INSIGHTS FROM VOTING, EMISSIONS, AND HEALTH POLICY
https://iuj.repo.nii.ac.jp/records/2000190
https://iuj.repo.nii.ac.jp/records/20001903df80ec6-6710-4081-98ab-2c00e308ca61
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
Exam Report_1C2D02_Han Thi Thanh Tam.pdf (210 KB)
|
|
|
|
Abstract_1C2D02_Han Thi Thanh Tam.pdf (101 KB)
|
|
|
|
Full Text_1C2D02_Han Thi Thanh Tam.pdf (1.7 MB)
|
|
| Item type | 学位論文 / Thesis or Dissertation(1) | |||||||
|---|---|---|---|---|---|---|---|---|
| 公開日 | 2025-09-02 | |||||||
| タイトル | ||||||||
| タイトル | APPLYING MACHINE LEARNING TO UNCOVER PATTERNS IN ECONOMIC ISSUES: INSIGHTS FROM VOTING, EMISSIONS, AND HEALTH POLICY | |||||||
| 言語 | en | |||||||
| 言語 | ||||||||
| 言語 | eng | |||||||
| 資源タイプ | ||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_46ec | |||||||
| 資源タイプ | thesis | |||||||
| 著者 |
Han Thi Thanh Tam
× Han Thi Thanh Tam
|
|||||||
| 著者（英） | ||||||||
| 姓名 | Han Thi Thanh Tam | |||||||
| 抄録 | ||||||||
| 内容記述タイプ | Abstract | |||||||
| 内容記述 | This dissertation applies Machine Learning (ML) to revisit three critical topics in socio-economic research including economic voting behavior, CO2 emissions forecasting, and the causal impact of health insurance reform. First, I examine how economic conditions shape electoral outcomes across 441 elections in 36 OECD countries (1960–2019). Voters tend to reward incumbents during growth and penalize them in downturns, with e!ects varying by context. Unemployment and Gross Domestic Product (GDP) growth are more influential during economic distress, while inflation is more salient in stable environments. The impact is stronger in countries with higher human capital and income, and most pronounced under left-leaning governments. Second, I revisit the Environmental Kuznets Curve and forecasts long-term CO2 emissions. Using data from 118 countries (1990–2018), I build a simple model based on GDP per capita and population. The results confirm an inverted- U pattern: emissions rise with income, then decline at higher income levels. Forecasts from 2019 to 2033 show slow emissions growth in high-income countries (0.02%–0.48% per year), but much faster growth in low-income countries (1.38%–2.20%). Finally, I estimate the impact of Social Health Insurance (SHI) on health outcomes in Vietnam using 2014 Vietnam Household Living Standard Survey data. To address endogeneity from adverse selection, the analysis applies a Doubly Robust Instrumental Variable (DRIV) approach. Results show that SHI significantly improves self-reported health. The e!ect is stronger among vulnerable groups, including low-income individuals, the unemployed, rural residents, and agricultural workers. Those with higher education also benefit more, likely due to better health literacy and ability to navigate the system. These findings highlight the need for targeted policies to enhance SHI e!ectiveness. This dissertation highlights how ML enhances empirical research by capturing complex patterns, improving prediction, and enabling precise causal analysis. These advances overcome traditional limitations and o!er valuable insights for policy design. The findings reinforce the role of data-driven approaches in governance, environmental planning, and public health. |
|||||||
| 学位名 | ||||||||
| 学位名 | 博士（経済学）/ Ph.D. in Economics | |||||||
| item_10006_degree_grantor_9 | ||||||||
| 学位授与機関名 | International University of Japan | |||||||
| 学位授与年度 | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | 2025 | |||||||
| 学位授与年月日 | ||||||||
| 学位授与年月日 | 2025-06-26 | |||||||
| dissertation_number | ||||||||
| 学位授与番号 | 経済博第2号 | |||||||
