The data used for the population database was derived from the United Nations web site.
UN excel data sheets (1950-2010)

The data was for all countries extending from 1950-2010 on an annual basis. It was grouped into five year age brackets (0-4),(5-9),(10-14)...etc.

Prior to 1990 all people over 80 years of age were grouped into at 80+ category. For the period 1990-2010 groups were extended to 90-100 with an 100+ group.

Given data in 5 year groups it is possible to derive death rates for each group.

To give an example...if we take the 5-10 year age group for the year 1960 they will be the same indiviuals who five years later (1965) will be the 10-15 year group. The numbers in this 10-15 year age group will be slightly less due to a small number of deaths. From the difference in numbers we can calculate a death rate for the 1960 5-10 year group.

The death rates thus calculated can be used to "predict the past" and by extrapolation forward of expected future death rate we can make future projections.

It is expected that death rates in the future will not alter much. I have used only "Low" and a "High" death rates. The "Low" rate is that for modern day France. The "High" that for modern day Bolivia.

Given data in 5 year groups it is NOT possible to calculate a birth rate. This would need to be obtained separately.

I have therefore used a NET birth rate (birth rate - death rate) for the 0-5 year group. This makes no difference to the projected population figures but does lead to a lower birth rate and to a slightly lesser death rate for the whole population.

See discussion item "Net vs actual birth rate" for detailed calculations of effect of this on projected birth and death rates.