Advances and Applications in Statistics
Volume 44, Issue 1, Pages 43 - 56
(January 2015) http://dx.doi.org/10.17654/ADASJan2015_043_056 |
|
ANALYSIS OF CANCER INCIDENCE DATA USING GENERALIZED LINEAR MODEL WITH NESTED EFFECT
Serpil AktaÅŸ
|
Abstract: This paper compares age-adjusted cancer incidence rates for the most common cancer types in the USA using a generalized linear model (GLM) with nested factors. This GLM methodology affords an opportunity to evaluate if one factor is nested within other factor. Nested effects are basically the main effects and the classical methods would not be the appropriate methods for the experiments where the levels of one factor are different depending on the levels of another factor. The data set utilized here is the form of nested factor, because cancer types are nested within the gender factor. GLM with nested factor enables us more appropriate solution and interpretation of gender specific cases. Data provided by SEER (USA: 2005-2009) were used for these analyses. The results indicated that prostate cancer is the most common malignancy among men in the USA, while carcinoma of the lung and bronchus are the most common cancer sites among American women. The results also imply that males are about 1.77 times more likely than females to get cancer in the USA. |
Keywords and phrases: cancer incidence rate, GLM, nested factor. |
|
Number of Downloads: 399 | Number of Views: 1305 |
|