A Multiple-State Model for Estimating the Probability of Cancer in Taiwan

• Main Authors: Chung-Ju Chiang, 蔣忠儒
• Other Authors: Lie-Fen Lin
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/27949503718789759799

碩士 === 逢甲大學 === 統計與精算所 === 100 === Cancer is the leading cause of death in Taiwan. It is the most important factor of the mortality rate. Insurance companies underestimate the cancer insurance prices and incur financial loss in cancer insurance. As the result, it is necessary to estimate reasonably the probability of getting cancer. In this paper, we use the multiple-state model to explore Taiwan’s medical data. The model divides people into three states: healthy, having cancer, and deceased. We then discuss the transition between these three states, as well as the intensity of this transition, with regard to a subject’s age, sickness duration, and gender. Generally speaking, we found that the mortality rate increased with the age of the subject, but between the age 20 and 50 there is a noticeable raise and decline in this rate. The age do not effect significantly the rehabilitation if someone has had cancer for a year Finally, we calculated the probability of going from healthy to sick in each year of a subject’s life. We found that the probability of a healthy subject getting cancer in a year peaks at age 63. The probability of getting cancer with females is more than with males between the age 28 and 63.