LCFIT -- Lee Carter FITter

LCFIT is a python program that implements the Lee-Carter mortality
model and forecasting algorithm.  Basically, the Lee-Carter model uses
the Singular Value Decomposition to decompose a matrix of mortality
rates into a single time series ("k_t") showing the overall mortality
trend, a base set of age specific mortality rates ("a_x"), and a
vector ("b_x") that describes the amount of mortality change at a give
age for a unit of overall mortality change.

This project uses mod_python + apache to serve the pages, postgresql
to store the data, scipy + numpy to do the calculations, and
matplotlib to make graphics.

This code repository also contains some code written in R and Matlab
to perform LCFIT in other contexts.

Many of the papers that were used to develop this code can be found on
Professor Ronald D. Lee's CEDA website:

http://www.ceda.berkeley.edu/peoplenew/rlee.html

This work was funded by CEDA as well.

An example of this code in action can be found at

http://lcfit.demog.berkeley.edu