在國際間的共識表述，肝炎治療以口服藥物為主，療程標準依達到表面抗原消失，視為療程之終點。此過程並不易達成，甚至可能為終生治療，但是在臨床上亦存在表現相對良好，數值下降較為迅速之患者，故此在本研究中使用 ARMA 模型建立估計及預測個人化療程，找出每位患者各自之趨勢，目的是探索是否存在某種共通之現象，以及增進療程中可運用之資訊。在本研究中，共整理 124 位患者，使用定量表面抗原作為研究資料，本研究依前人文獻所示，採用是否到達定量表面抗原達 100IU/mL 進行患者間分類，這是由於文獻表明當患者達到此數值後有較高之機會痊癒，故此在本研究依照是否達到定量表面抗原 100IU/mL 分為三種類型之患者，A 類型患者、B 類型患者、C 類型患者，再依估計之趨勢進行更為細緻的分類，整理可能之現象並加以統計。 統計分析後，本研究歸納出維持定量表面抗原 100IU/mL 內的 A 類型患者無趨勢以及緩趨勢的比例各佔50%；曾經達到過定量表面抗原 100IU/mL 的 B 類型患者則存在四種類型趨勢，分別為 aa 之兩段皆無趨勢、ca 之前段趨勢強後段無趨勢、cb 之前段趨勢強後段趨勢弱、cc 之前後段皆強，最後則是未達到定量表面抗原 100IU/mL 的 C 類型患者，分為無趨勢以及強烈下降趨勢各為 50%。根據本研
究的歸納，整理出當患者達到定量表面抗原 100IU/mL 後，以 B 類型患者趨勢 cb及 cc 以及 A 類型患者趨勢 b，預測有望達到表面抗原消失之比例有四成左右，亦歸納出有五成以上之患者在定量表面抗原達到 100IU/mL 後，並不存在趨勢，預測可能為終生治療。

The international consensus stated that hepatitis B treatment is mainly oral drugs. The standard of the course of treatment is based on the disappearance of the hepatitis B surface antigen (HBsAg), which is regarded as the end of treatment. But it's not easy to achieve, and could even be a lifetime treatment, but there are also patients with better performance about rapid decline in HBsAg. Therefore, in this study, the ARMA model was used to estimate and forecast the individual course and to find out each patient's own trend in treatment, whether there is a common phenomenon and to improve the information that can be used in the treatment to against hepatitis B viral.

In this study, a total of 124 patients were used as the research data. Most patients with hepatitis B virus could undetectable in treatment, but patients did not recover at this time, because the hepatitis B surface antigen was produced by hepatitis B viral , so when the hepatitis B surface antigen disappears, it finally can be said that the treatment is over and recovery of patients. In previous literature, these study use quantitative
hepatitis B surface antigen of under 100IU/mL to shows when the patients reaches under 100IU/mL have a higher chance of recovery. Because previous literature in this study patients were divided into three types analysis, analysis A patients, analysis B patients, and analysis C patients, then classify them in more detail with the estimated trend, at last sort out possible phenomenon.

After study, this research concluded that, the analysis A patients who maintained the quantitative hepatitis B surface antigen within 100IU/mL had no trend and a slow decline trend. Then there are four types of trend in analysis B patients who had reached a quantitative hepatitis B surface antigen of 100IU/mL. Finally, two types of trend in analysis C patients who hadn't reached the quantitative hepatitis B surface antigen
100IU/mL.

In this study conclusion, it is expected that the proportion of quantitative hepatitis B surface antigen disappearance is about 45%, and more than 50% of patients may be treated for whole life.

目錄
論文審定書 ........................................................................................................................i
致謝 ................................................................................................................................... ii
摘要 .................................................................................................................................. iii
Abstract ............................................................................................................................. iv
目錄 ................................................................................................................................... v
圖目錄 .............................................................................................................................. vi
表目錄 ............................................................................................................................ viii
第一章 緒論 .................................................................................................................... 1
第一節 研究動機 .................................................................................................... 1
第二節 本文架構 .................................................................................................... 2
第二章 文獻回顧 ............................................................................................................ 3
第三章 研究方法 .......................................................................................................... 10
第一節 ARMA 估計模型 ......................................................................................... 10
第二節 離群值處理 .............................................................................................. 12
第三節 填補遺失值 .............................................................................................. 16
第四節 AIC & BIC 配適度評估 ........................................................................... 18
第五節 ACF & PACF 判斷模型 ............................................................................. 19
第六節 模型殘差檢定 .......................................................................................... 21
第七節 預測 .......................................................................................................... 23
第四章 統計實驗結果 .................................................................................................. 25
第一節 詴驗患者分類 .......................................................................................... 25
第二節 A 類型患者分析 ....................................................................................... 25
第三節 B 類型患者分析 ....................................................................................... 32
第四節 C 類型患者分析 ....................................................................................... 53
第五章 結論與限制 ...................................................................................................... 62
第一節 研究結論 .................................................................................................. 62
第二節 研究建議 .................................................................................................. 64
第三節 研究限制 .................................................................................................. 64
參考文獻 ........................................................................................................................ 66

王姿乃, 于明暉, 廖運範, 林燈寅, 陳建仁. (1994). “B 型肝炎表面抗原帶原者肝硬化之多重危險因子研究.” 中華公共衛生雜誌, 13(3), 258-268.
李慶男, (2006). “時間序列分析.” 國立中山大學經濟學研究所.
陳旭昇, (2009). “時間序列分析,總體經濟與財務金融之應用. ”東華書局出版.
陳國樑, 楊佩烜, 黃勢彰. (2020). “台灣稅收預測表現之探討.” 人文及社會科學集刊 32(2), 271-319.
楊踐為, 李家豪, 類惠貞. (2007). “應用時間序列分析法建構台灣證券市場之預測交易模型.” 中華管理評論國際學報, vol. 10, No.3
蘇苗彬 與 陳旺志. (1996). “臺北市山坡地降雨量及地下水位之時間序列分析.” 中華水土保持學報, 27(2), 127-138.
蘇彥同. (2017). “以資料探勘技術探討 B 型肝炎治療前後之影響因素--以某公立醫院資料為分析案例.” 國立台北科技大學.
Akaike, H. (1974). “A new look at the statistical model identification.” IEEE transactions on automatic control, 19(6), 716-723.
Ahmed, M. S., & Cook, A. R. (1979). “Analysis of freeway traffic time-series data by using Box-Jenkins techniques. ” (No. 722).
Ariyo, A. A., Adewumi, A. O., & Ayo, C. K. (2014, March). “Stock price prediction using the ARIMA model. ” In 2014 UKSim-AMSS 16th International Conference on Computer Modelling and Simulation (pp. 106-112). IEEE.
Alsharif, M. H., Younes, M. K., & Kim, J. (2019). “Time series ARIMA model for prediction of daily and monthly average global solar radiation: The case study of Seoul, South Korea.” Symmetry, 11(2), 240.
Blumberg, B. S., & Alter, H. J. (1965). “A new antigen in leukemia sera.” JAMA, 191(7), 541-546.
Box, G. E., Jenkins, G. M., & Reinsel, G. (1970). “Time series analysis: forecasting and control Holden-day San Francisco.” BoxTime Series Analysis: Forecasting and Control Holden Day1970.
Bollerslev, T. (1986). “Generalized autoregressive conditional heteroskedasticity. ”Journal of econometrics, 31(3), 307-327.
Chang, I., Tiao, G. C., & Chen, C. (1988). “Estimation of time series parameters in the presence of outliers.” Technometrics, 30(2), 193-204.
Chen, Y. C., Huang, S. F., Chu, C. M., & Liaw, Y. F. (2012). “Serial HBV DNA levels in patients with persistently normal transaminase over 10 years following spontaneous HBeAg seroconversion.” Journal of viral hepatitis, 19(2), 138-146.
Chevaliez, S., Hézode, C., Bahrami, S., Grare, M., & Pawlotsky, J. M (2013). “Long-term hepatitis B surface antigen (HBsAg) kinetics during nucleoside /nucleotide analogue therapy: finite treatment duration unlikely.” Journal of hepatology, 58(4), 676-683.
Dickey, D. A., & Fuller, W. A. (1979). “Distribution of the estimators for autoregressive time series with a unit root. ” Journal of the American statistical association, 74(366a), 427-431.
Engle, R. F. (1982). “Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. ” Econometrica: Journal of the Econometric Society, 987-1007.
Engle, R. F., & Granger, C. W. (1987). “Co-integration and error correction: representation, estimation, and testing. ” Econometrica: journal of the Econometric Society, 251-276.
Granger, C. W., Newbold, P., & Econom, J. (1974). “Spurious regressions in econometrics. ” Baltagi, Badi H. A Companion of Theoretical Econometrics, 557-61.
Hamilton, J. D. (1994). “Time Series Analysis. ” Princeton, NJ: Princeton University Press.
Hyndman, R. J., & Athanasopoulos, G. (2018). “Forecasting: principles and practice.” OTexts.
Jarque, C. M., & Bera, A. K. (1987). “A test for normality of observations and regression residuals.” International Statistical Review/Revue Internationale de Statistique, 163-172.
Ljung, G. M., & Box, G. E. (1978). “On a measure of lack of fit in time series models.” Biometrika, 65(2), 297-303.
McKinley, S., & Levine, M. (1998). “Cubic spline interpolation.” College of the Redwoods, 45(1), 1049-1060.
Mills, T. C. (2012). “A very British affair: Six Britons and the development of time series analysis during the 20th century. ” Palgrave Macmillan.
Moritz, S., & Bartz-Beielstein, T. (2017). “imputeTS: time series missing value imputation in R.” R J., 9(1), 207.
Nelson, C. R., & Plosser, C. R. (1982). “Trends and random walks in macroeconmic time series: some evidence and implications. ” Journal of monetary economics, 10(2), 139-162.
Raymond, Y. C. (1997). “An application of the ARIMA model to real‐estate prices in Hong Kong. ” Journal of Property Finance.
Slutzky, E. (1937). “The summation of random causes as the source of cyclic processes. ” Econometrica: Journal of the Econometric Society, 105-146.
Schwarz, G. (1978). “Estimating the dimension of a model. ” The annals of statistics, 6(2), 461-464.
Sims, C. A. (1980). “Macroeconomics and reality. ” Econometrica: journal of the Econometric Society, 1-48.
Stock, J. H., & Watson, M. W. (1988). “Testing for common trends. ” Journal of the American statistical Association, 83(404), 1097-1107.
Tseng, T. C., Liu, C. J., Su, T. H., Wang, C. C., Chen, C. L., Chen, P. J., & Kao, J. H. (2011). “Serum hepatitis B surface antigen levels predict surface antigen loss in hepatitis B e antigen seroconverters. ” Gastroenterology, 141(2), 517-525.
Wold, H. (1938). “A study in the analysis of stationary time series ”(Doctoral dissertation, Almqvist & Wiksell).
Whittle, P. (1951). “Hypothesis testing in time series analysis (Vol. 4). ” Almqvist & Wiksells boktr..
Yule, G. U. (1927). “VII. On a method of investigating periodicities disturbed series, with special reference to Wolfer's sunspot numbers. ” Philosophical Transactions of the Royal Society of London. Series A, Containing Papers of a Mathematical or Physical Character, 226(636-646), 267-298.