在競爭風險下，區間設限資料的門檻模型

First-Hitting-Time Models for Competing Risks with Interval-Censored Data

洪嘉彣(Chia-Wen Hung)

指導教授 ： 陳蔓樺

淡江大學/商管學院/統計學系應用統計學碩士班/碩士(2022年)

https://doi.org/10.6846/TKU.2022.00024

摘要

近年來，隨著醫療科技日益進步，相關研究亦與日俱增。一般臨床實務上通常無法得知病患發生感興趣事件的確切時間，因此大多收集到的資料為設限資料。而在追蹤病患的過程中，並非只關注病患經歷單一事件，其病程中可能存在多個潛在事件。當病患先經歷其他（非感興趣）事件，使得我們無法觀察到感興趣事件發生，此時有競爭風險問題，是需要進一步探討。若不考慮競爭風險問題，可能使估計產生偏差。
本篇考慮在競爭風險情況下，針對區間設限資料使用門檻模型，並結合EM演算法與牛頓迭代法估計參數，使用費雪訊息矩陣估計標準誤。

關鍵字

區間設限 ； 競爭風險 ； 門檻模型 ； EM演算法 ； 牛頓迭代法 ； 費雪訊息矩陣

並列摘要

With the progress of medical technology, there are growing related research in recent years. In general clinical practice, it is hard to observe the exact time of the event happened, we usually know it happened in a particular time interval. In the process of tracing the patients, it is not only concerned about the patients’ experience of a single event. There may be multiple potential events in the course of the disease. When the patients experience other events first, we can’t observe the occurrence of the event of interest, then there is a problem of competing risk. Without considering the issue of competing risk, the estimate may be biased.
Considering the issue of competing risk, we use first-hitting-time models to analyze the censored data. With EM algorithm and Newton-Raphson method to estimate the parameter and Fisher information matrix to estimate the standard error.

並列關鍵字

Censored sample ； Competing risk ； First-hitting-time model ； EM algorithm ； Newton-Raphson method ； Fisher information matrix

參考文獻

[1] Berkson, J. and Gage, R. P. (1952). Survival Cure for Cancer Patients Following Treatment. Journal of the American Statistical Association, 47, 501-515.

[2] Boag, J. W. (1949). Maximum Likelihood Estimates of the Proportion of Patients Cured by Cancer Therapy. Journal of the Royal Statistical Society, Series B, 11, 15-53.

[3] Farewell, V. T. (1992). The Use of Mixture Models for the Analysis of Survival Data with Long-Term Survivors. Biometrics, 38, 1041-1046.

[4] Fine, J. P. and Gray, R. J. (1999). A Proportional Hazards Model for the Subdistribution of a Competing Risk. Journal of the American Statistical Association, 94, 496-509.

[5] Greenhouse, J. B. and Wolfe, R. A. (1984). A competing risks derivation of a mixture model for the analysis of survival data. Communications in Statistics - Theory and Methods, 13, 3133-3154.