The LIFE-M data, which has been more than seven years in the making, have been released. The data can be downloaded from OPEN ICPSR here.
The LIFE-M project combines millions of U.S. vital records (birth, marriage, death certificates) with census information into a longitudinal and intergenerational micro-database. With the help of cutting-edge, machine learning techniques, the LIFE-M data follow four generations of Americans from birth to death. High quality training data is used to achieve large-scale performance at high rates of precision. Birth cohorts begin in late 1800s and include their great grandchildren born between 1915 and 1975.