簡易檢索 / 詳目顯示
| 研究生： |
陳孟翔 Chen, Meng-Hsiang |
|---|---|
| 論文名稱： |
利用整數值時間序列模型監控登革熱爆發時間點 Monitoring outbreak of Dengue Fever by using INARMA model |
| 指導教授： |
林良靖
Lin, Liang-Ching |
| 學位類別： |
碩士 Master |
| 系所名稱： |
管理學院 - 統計學系 Department of Statistics |
| 論文出版年： | 2016 |
| 畢業學年度： | 104 |
| 語文別： | 英文 |
| 論文頁數： | 28 |
| 中文關鍵詞： | 整數值時間序列 、蒙地卡羅馬可夫鏈 、登革熱 |
| 外文關鍵詞： | Integer-valued time series model, Markov Chain Monte Carlo, Dengue fever |
| 相關次數： | 點閱：78 下載：5 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
登革熱已成為全球性的問題，而台灣近三年夏季的登革熱病例人數也逐年上升。衛生局訂定疫情失控的標準為一個縣市單日病例數達30 人，而本研究目的是建立新的監控標準，期望能更早監測出登革熱疫情失控的時間點。給定一組平穩的整數值時間序列資料，本研究利用Neal and Rao (2007) 提出的蒙地卡羅馬可夫鏈方法估計參數與預測值的後驗分佈。接著利用每個預測值後驗分佈的樣本百分位數建立管制界線，再以西方電氣法則來監控失控的時間點。實證上以高雄2012 年登革熱每日發病人數為基準，建立管制圖來監控高雄2013 到2015 和台南2014 到2015，共五個年度的登革熱疫情。其中，高雄2013 年和台南2014 年屬於疫情穩定，本研究建立的管制圖與衛生局監控相同，並沒有出現過度反應的情況。而高雄2014、2015 年與台南2015 年為疫情失控的年度。與衛生局的監控標準相比，本研究建立的管制圖能提早監測出登革熱疫情失控。
Dengue fever has become a global issue. The case of dengue fever is increasing in recent three years. Public Health Bureau (PHB) detects the uncontrolled epidemic in the case of 30 people inflected by the dengue fever per day. The purpose of this study is to construct a new criterion in order to detect the outbreak timing earlier. To begin with, given a stationary integer-valued time series observations, the MCMC algorithm proposed by Neal and Rao (2007) can be used to estimate the posterior distribution for parameter and predictive value. For each predictive value, the control limits is constructed by the sample quantiles of the posterior distributions. Then, the out-of-control points are detected by the first four Western electric rules. In real data analysis, we consider the inflection number of daily dengue fever
in Kaohsiung 2012 to be the benchmark. We detect the outbreak of dengue fever on Kaohsiung (from 2013 to 2015) and Tainan (from 2014 to 2015). In Kaohsiung 2013 and Tainan 2014, there are no uncontrolled epidemic and our control chart shows no out-of-control point as well as PHB’s criterion. In Kaohsiung 2014 and 2015 and Tainan 2015, the outbreak times of our control chart are detected earlier than PHB’s criterion.
1. Al-Osh, M. A., and Alzaid, A. A. (1987). First-order integer-valued autoregressive (INAR (1)) process. Journal of Time Series Analysis, 8(3), 261-275.
2. Cardinal, M., Roy, R., and Lambert, J. (1999). On the application of integer-valued time series models for the analysis of disease incidence. Statistics in Medicine, 18(15), 2025-2039.
3. Dion, J. P., Gauthier, G., and Latour, A. (1995). Branching processes with immigration and integer-valued time series. Serdica Mathematical Journal, 21(2), 123-136.
4. Freeland, R. K., and McCabe, B. P. (2004). Analysis of low count time series data by Poisson autoregression. Journal of Time Series Analysis, 25(5), 701-722.
5. Hsieh, Y. H., and Chen, C. W. S. (2009). Turning points, reproduction number, and impact of climatological events for multi-wave dengue outbreaks. Tropical Medicine
& International Health, 14(6), 628-638.
6. McKenzie, E. (2003). Ch. 16. Discrete variate time series. Handbook of statistics, 21, 573-606.
7. Neal, P., and Subba Rao, T. (2007). MCMC for integer-valued ARMA processes. Journal of Time Series Analysis, 28(1), 92-110.
8. Steutel, F. W., and Van Harn, K. (1979). Discrete analogues of self-decomposability and stability. The Annals of Probability, 7(5), 893-899.
校內：2021-07-20公開校外：2021-07-20公開