# Getting MLEs of a Denny's model survival curve

Last updated on 6 January 2012 (Fri)

## References

1. Denny, C. (1997) A model of the probability of survival from birth. Mathematical and Computer Modelling, 26: 69-78.
2. Kobayashi, K. and Z. Nanjo (1988) Generation life tables for Japan based on period life tables covering the years 1891-1986. Nihon University Population Research Institute, Tokyo.
3. Nelder, J.A. and R. Mead (1965) A simplex method for function minimization. The Computer Journal, 7(4): 308-313.

## ダウンロード

dennys.exe (Win32自己解凍ファイル，27966 bytes)，または，dennys.tgz (tar+gzip圧縮ファイル，9055 bytes)のどちらかをダウンロードしてください。

## 著作権等

revision 1.0 All right reserved by Minato Nakazawa <minato-nakazawa[atmark]umin.net>.

## 配布条件

When extracting, you may get denny.txt (explanation like this page), denny.c (the source file written in C, can be compiled by gcc), denny.exe (Win32 executable file, compiled by Mingw32), jpcohortm.dat (sample data based on the cohort life table of Japanese males, Kobayashi and Nanjo, 1988).

## Description

The program estimates 3 parameters of Denny's model for survival curve (Denny, 1997), with best fit to any given data. Input data file (survival curve) must have age x, <tab>, survival rate l(x), which are aligned in each line and the lines compose whole survival curve from age 0 to eldest age (see, sample data file, jpcohortm.dat).

When running the program, if the data file is given as 1st command line parameter, then the best fit parameters will be shown in standard output, using so-called downhill simplex method for function minimization to achieve minimum root MSE (Nelder and Mead, 1965). If obtained RMSE is not small probably due to local minimum pocket, please change initial parameter values by giving 2nd, 3rd, and fourth command line parameters.