即時變化點檢測是辨別序列數據是否隨著時間的推移而發生結構變化的過 程。在實務上，相關性的結構是時間序列分析的重要問題。此外，為了放寬相 關性的限制，我們提出了一個建立在 Clayton copula 且其邊際分布為常態分佈的 copula-馬可夫模型，並將我們提出的模型在不同的情況下與獨立的模型以及一 階自我迴歸的模型進行比較。模擬的結果指出無論在何種情況下，我們提出的 模型在準確率以及平均絕對誤差下皆表現得比其他兩個模型來的好。在實證研 究中，我們考慮且偵測標準普爾 500 指數、日經 225 指數和富時 100 指數的每 日對數報酬率在 2008 金融危機和 2020 冠狀病毒疾病大流行下報酬率的變化點， 實證結果揭露我們提出的模型是可以捕捉有序列相關資料的結構變化。;Online changepoint detection is a procedure to identify whether a sequential data structure changes over time. In practice, the dependent structure is an important issue for time series analysis. To achieve flexibility limit dependence, we propose a copulabased Markov model based on the Clayton copula and the marginal distribution being a normal distribution and compare the proposed model with the independent model and the first-order autoregressive model under various scenarios. The simulation results indicate that the proposed model outperforms the other models in precision and mean absolute error (MAE) no matter the scenarios. For empirical studies, we consider the daily log returns of the S&P 500 Index, the Nikkei 225 Index, and the FTSE 100 Index to identify the changepoints in the period of the financial crisis in 2008 and the COVID-19 pandemic in 2020. Results reveal that the proposed model is able to capture the structure change for serial dependent data
