精確的癌症死亡率地圖可以有效表現死亡率的空間變異。本研究目的即在利用標 準化死亡比值，透過局部平滑技巧得出較穩定的新統計量，並根據此統計值繪製臺灣癌症地 圖。利用傳統標準化死亡比值所繪製之地圖，在稀有癌症種類或人口數少之鄉鎮市區，常因 死亡率估計值變異過大而不穩定。利用模擬方法，比較本文所提統計量與標準化死亡比值之 穩定性大小。結果發現，不論對稀有癌症種類或人口數少之鄉鎮市區所算出之估計值，本文 所提之統計量都有較低的變異量。

　　　　　The cancer map is a useful tool to show the spatial variation of mortality rate. Hence, it is necessary to make the map with high resolution. The map for the case of rare disease or small population size is unstable when plotted by the conventional standardized mortality ratio (SMR). The purpose of this paper is to provide a new estimator applying the smoothing technique to the conventional SMR. A simulation study is presented to compare the difference between the new estimator and the conventional SMR. It is found that the proposed estimator has a substantial improvement for stabilizing variance in either case of rare cancer disease or small population size.