Quantifying demographic and socioeconomic transitions for computational epidemiology: an open-source modeling approach applied to India

Table 2 Comparisons of three calibrated models reveal that one incorporating secular trends is more consistent with the observed data^a

Model	Components	ΔDIC when fit against Table 1 data sources
1	Age, sex, urban/rural residence, fertility, mortality	Reference
2	Age, sex, urban/rural residence, fertility, mortality, educational attainment	+5.2 versus model 1
3	Age, sex, urban/rural residence, fertility, mortality, educational attainment, migration	−259.1 versus model 2

^a A model incorporating both education and migration rates best explains the variance in the data, even when penalizing the use of more parameters using the deviance information criterion (DIC). Note that lower DIC scores are considered better (reflecting better fit to data and fewer parameters to accomplish the fitting), and a >10 point difference is considered meaningful [42]

ISSN: 1478-7954