Data originally analysed by Andersen et al (1993) on the survival of 205 patients in Denmark with malignant melanoma, and used by Claeskens and Hjort (2008) to illustrate focused model selection in Cox regression.
melanomaA data frame with 205 rows and the following columns:
Patient identification number
Survival status: 1 = dead from illness, 2 = censored, 4 = dead from other causes
Survival time in days
Invasion depth: factor with levels 1, 2, 3
Infection infiltration level, a measure of resistance to the tumour: factor with levels 1 (high resistance), 2, 3, 4 (low resistance)
Indicator for epithelioid cells present
Indicator for ulceration
Thickness of the tumour in 1/100 mm
Sex. Factor with levels "female","male" and reference level "female"
Age in years
Survival time in years (instead of days)
Version of thick centred around its mean and rescaled, defined as (thick - 292)/100.
Claeskens, G., & Hjort, N. L. (2008). Model selection and model averaging (Vol. 330). Cambridge: Cambridge University Press.
Andersen, P. K., Borgan, O., Gill, R. D., & Keiding, N. (2012). Statistical models based on counting processes. Springer.