Transcript for estimating the number of population

Methodology of estimating the
annual number of usual resident
population in Latvia
Baiba Zukula
Deputy Director of Social Statistics Department
Central Statistical Bureau of Latvia
Content
1. Introduction
2. History of population estimations
3. New methodology of calculation of
the number of population (since
2012)
4. The estimated number of
population (at the beginning of
2016)
5. Conclusions
2
LATVIA
- Independent country since 1991;
- EU country since 2004 (euro since
2014);
- Number of population less than 2 mln
- Territory – 64 m2
3
Number of population in Latvia
(1950-2016)
- 26 %
4
Population in Latvia at the
beginning of 1990 and 2016
(by age and gender)
5
History of population estimations
(I)
Calculation of the number of population before
the Population Census 2011:
Data from the Population Register of
the Office of Citizenship and Migration
Affairs (OCMA) – monthly data
received by CSB;
Data from the Registry Offices – data
on born or dead at the end of the year
had to be added
• registration of birth – one month
after birth allowed
• registration of death – 6 working
days allowed
6
History of population estimations
(II)
Population Census 2011 allowed to update the
number of population at the beginning of 2011
- 2 mln 75 thsd.
The difference between Census and Population
Register data - 155 thsd or about 7 % of the
resident Latvian population registered in the
Population Register.
Reason - unregistered migration – according
to the Population Register, usual residence of
person – Latvia, however, according to the
Census data – outside Latvia.
7
It was ascertained, that it is necessary
to find out the ways how to estimate
more
precisely
the
number
of
population.
The possible ways:
 To improve the data quality on
usual residence in the Population
Register;
 To evaluate methods applied by
other countries;
 To work out a new method.
8
New methodology of calculation
of the number of population (I)
In 2012 the CSB worked out a new
method for estimating the number of
population.
The method is based on:
• The Population Register,
• Other administrative register data.
The aim of the method – to estimate the
status of the actual or de facto place of
residence at the beginning of the year for
every resident registered (on individual
level) in Latvia.
9
New methodology of calculation
of the number of population (II)
CSB has access to data from following administrative
registers:
• State Revenue Service (SRS);
• State Social Insurance Agency (SSIA);
• Ministry of Education and Science (MES);
• Agricultural Data Centre (ADC);
• Rural Support Service (LRSS);
• National Health Service (NHS);
• State Employment Agency (SEA);
• Road Traffic Safety Directorate;
• Data on persons in imprisonment, in social care
institutions;
• Information from local governments on social
benefits paid;
• Biggest higher education institutions.
1
0
Certain conditions had to be taken into
account:
• Estimation of the total number of
population and also the international
migration (immigration and emigration)
is needed;
• Data on population is necessary in small
groups (both for national and
international data users):
- by sex and age (one year age group);
- by territorial units by three main age
groups;
- by regions by one year age group;
- international long term emigration and
immigration by sex, one year
age group etc.
1
1
Calculation method (I)
• The aim of the model is to evaluate the de
facto residence status of each individual
registered in Latvia.
• The Logistic regression model had been
developed to evaluate the residence status
(model
developed
by
Mathematical
Support Division of CSB – Jeļena
Vaļkovska, Mārtiņš Liberts).
• With the help of administrative register
data, on each person registered within the
Population Register there are more than
200 characteristics variables developed.
1
2
Calculation method (II)
The aim of the model is to predict the
probability of being a resident for each
individual.
Important – if the information about the
person is in any register, it doesn’t mean that
the person will be counted as usual resident.
It just increase the probability!
The model had been developed using data
from 2011 Population Census (1 March 2011)
on actual place of residence and data from
administrative data sources on 2010, 1
January 2011 or 1 March 2011 (thus there
should be no break down in series).
1
3
The number of population at the
beginning of 2016
(Comparison of the Population Register (PR)
and CSB estimation)
1
4
Evaluation of the preciseness of
estimation of the number of
population
Comparison
of
the
prepared
evaluation with:
• respondents in CSB surveys
(LFS, EU-SILC, EHSIS)
In total we compared more than 200
thousand persons (10% of total
number
of
population)
at
the
beginning of 2015. Only about 1%
were not included in the estimated
number of population.
1
5
Conclusions (I)
The evaluated number of population – official
number of population in Latvia - is the only
one used for national and international
purposes.
The CSB continues the work to
preciseness of the worked out model.
improve
New individual data sources:
- Personalized public transport tickets (used in
the capital city);
- Individual data from libraries;
- Individual data on those who started studies
abroad and received student loans for studies
or living.
1
6
Conclusions (II)
To evaluate the accuracy of the method
Micro Census 2015 will be used.
Probably the Micro Census 2015 will be used
to update the model (model was developed
based on 2011 Census).
1
7
Thank you!
CSB methodology is available:
http://www.csb.gov.lv/en/statistikas-temas/metodologija/numberpopulation-and-key-demographic-indicators-36808.html
(look for pdf file
http://www.csb.gov.lv/sites/default/files/dati/demstat_metodologija_eng
_gala.pdf)