Scribd
Upload
42 views13 pages

Section 8 - Population Estimates and Projections

Demography studies population estimation and projections

Uploaded by

mashiringwaninyasha2003
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
42 views13 pages

Section 8 - Population Estimates and Projections

Demography studies population estimation and projections

Uploaded by

mashiringwaninyasha2003
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Section 8

Population Estimates and


Projections

ED 2021
 We consider three types of population
estimates
1) Intercensal estimates – these are estimates between 2 censuses.
2) Immediate postcensal estimate – these are estimates into the near
future after a census.
3) Long-term future estimates – these are usually called the
projections (when estimates periods for more than 5 years into the
future).
Inter and Postcensal Estimates
Given two or more population totals. It is reasonable to fit some
mathematical curve for the data in order to obtain various estimates.

Care is needed in choosing a sensible mathematical model and an


interpretation especially with estimates well into the future that is at
least 5 years.
Mathematical Models
1. Linear Interpolation It is the simplest method and based for/on small time
intervals and between censuses. E.g. between 𝑡1 and then 𝑡2 interpolate. In
real life observations do not usually grow linear. The linear model is

𝒕
𝑷𝒕 = 𝑷𝟎 + (𝑷𝒏 − 𝑷𝟎 )
𝒏

where 𝑃0 ~𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑏𝑎𝑠𝑒 𝑦𝑒𝑎𝑟 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛


𝑃𝑛 ~𝑠𝑒𝑐𝑜𝑛𝑑 𝑐𝑒𝑛𝑠𝑢𝑠 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑛 𝑦𝑒𝑎𝑟𝑠 𝑙𝑎𝑡𝑒𝑟
𝑃𝑡 ~𝑖𝑠 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑠𝑖𝑧𝑒 𝑡 𝑦𝑒𝑎𝑟𝑠 𝑙𝑎𝑡𝑒𝑟
𝑛~𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑐𝑒𝑛𝑠𝑢𝑠 (𝑛𝑜 𝑜𝑓 𝑦𝑒𝑎𝑟𝑠)
Example

Given the population of Zimbabwe is 1982 was 7.8million and


10.4million in 1992. Calculate the population in 1989.

Solution
Discuss in groups
Solution

𝑃0 = 𝑃82 = 7.8 𝑃1 = 𝑃92 = 10.4 𝑃89 = 𝑃7

7
𝑃89 = 𝑃0 = 7.8 + 10.4 − 7.8 = 9.62 𝑚𝑖𝑙𝑙𝑖𝑜𝑛
10

Homework
Discuss the advantages and disadvantages of this method.
2. Polynomial Model
Given three or more census points we can fit a quadratic or higher
order curve (by regression or simultaneous equations).
E.g. 𝑷𝒕 = 𝒂 + 𝒃𝒕 + 𝒄𝒕𝟐
Where 𝑎, 𝑏, 𝑐 𝑎𝑟𝑒 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡𝑠.

This is a superior method than linear models but very unreliable for
projections.
3. Geometric Model
This assumes a constant annual growth rate and is given by
𝒕
𝑷𝒕 = 𝑷𝟎 (𝟏 + 𝒓)
Where 𝑃0 ~𝑏𝑎𝑠𝑒 𝑦𝑒𝑎𝑟 𝑝𝑜𝑝
𝑃𝑡 ~𝑝𝑜𝑝 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡
𝑟~𝑖𝑠 𝑡ℎ𝑒 𝑔𝑟𝑜𝑤𝑡ℎ 𝑟𝑎𝑡𝑒
It provides satisfactory estimates for intercensal and immediate postcensal
periods.
Care should be taken using it for future projections because growth rate
does not normally remains constant for long periods especially in
developing countries.
In reality growth occurs continuously throughout the whole years. It may
be better to use the exponential model.
4. Exponential Model
It is given by 𝑷𝒕 = 𝑷𝟎 𝒆𝒓𝒕
Where 𝑃0 ~𝑏𝑎𝑠𝑒 𝑦𝑒𝑎𝑟 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛
𝑟~𝑖𝑠 𝑎 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 𝑎𝑛𝑑 𝑐𝑜𝑛𝑡𝑖𝑛𝑜𝑢𝑠 𝑔𝑟𝑜𝑤𝑡ℎ 𝑟𝑎𝑡𝑒
𝑃𝑡 ~𝑝𝑜𝑝 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡

When the growth rate 𝒓 is small, the value from the geometric model
and that from the exponential model are the same
Logistic Model
Any country can sustain a certain population size.
All previous methods assume that the population can grow out without a limit. A
logistic model is more sophisticated and assumes that there is a limit to a
population (Ultimate population)
𝒂
𝑷𝒕 =
𝟏 + 𝒃𝒆−𝒄𝒕
 Where a, b and c are constants
 𝑎~𝑖𝑠 𝑡ℎ𝑒 𝑢𝑙𝑡𝑖𝑚𝑎𝑡𝑒 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛
This is better than the previous method in many situations but since 3
parameters require 3 censuses
it is difficult to fit and not good for too long periods

Various modifications can be made to the logistic curve to achieve to the logistic
curve to achieve better conformity with the observed data.
The Component Method
If a country have an efficient vital registration system, the component
method is applicable.
Recall the balancing equation
𝑷𝒕 = 𝑷𝟎 + 𝑩 − 𝑫 + (𝑰 − 𝑬)
Where 𝑃𝑡 ~𝑝𝑜𝑝 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡
𝑃0 ~𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑝𝑜𝑝
𝐵 − 𝐷 ~𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑖𝑛𝑐𝑟𝑒𝑎𝑠𝑒
𝐼 − 𝐸 ~𝑛𝑒𝑡 𝑚𝑖𝑔𝑟𝑎𝑡𝑖𝑜𝑛 𝑟𝑎𝑡𝑒
Component method for estimation applies if vital events are accurate to
update census figures.
This is the most accurate and reliable method but requires considerable
accurate data.
Advantages
1) It enables one to appreciate the effect of each of the separate
assumptions about fertility, mortality and migration.
2) Projected values are available by age and sex distribution.

Disadvantages
Homework
Research on the Leslie’s Matrix
Thank you

The End

You might also like