Long-term migration of population

Data on long-term migration provide information on the number of persons arriving to the country for permanent residence or for at least one year, the number of persons departing from the country for permanent residence or for at least one year, as well as data on internal long-term migration.

N/A

N/A

N/A

Statistics database

EU data on Eurostat website sections: Immigration and Emigration.

N/A

N/A

In 2012, CSB worked out methodology that is based on statistical classification and used for the production of population statistics. The statistical classification is aimed at dividing population registered in Latvia into two classes (groups) – those actually living in Latvia (usually resident population of Latvia) and those actually living abroad. The model of statistical classification was developed by applying logistic regression analysis.

To make population estimate CSB uses the following administrative data sources:

• Office of Citizenship and Migration Affairs (OCMA);
• State Revenue Service (SRS);
• State Social Insurance Agency (SSIA);
• Ministry of Education and Science (MES);
• Agricultural Data Centre (ADC);
• Rural Support Service (RSS);
• National Health Service (NHS) (data on vaccination or testing for Covid-19 are not taken into account);
• State Employment Agency (SEA);
• Road Traffic Safety Directorate;
• Latvian Prison Administration;
• the Ministry of Welfare;
• State Chancellery.

Quality control of the estimate is ensured by using the following data:

• Social Assistance Administration Information System data on number of persons that have received local government allowances and services;
• individual data of the Administration of Studies and Research about persons who have received a study/ student loan for studies abroad (until 2021) and State Education Development Agency (since 2023);
• individual data of the Riga municipal limited liability company "Rīgas satiksme" about personalised e-tickets;
• information on persons who have participated in the largest household surveys.

N/A

In 2012, CSB worked out methodology that is based on statistical classification and used for the production of population statistics. The statistical classification is aimed at dividing population registered in Latvia into two classes (groups) – those actually living in Latvia (usually resident population of Latvia) and those actually living abroad. The model of statistical classification was developed by applying logistic regression analysis.

Description of method in PDF

Number of population from 1 January 2001 to 1 January 2011 has been re-calculated according to the results of Population and Housing Census 2011, taking into account non-registered emigration.

Since 2020, new indicators have been published - population by year of arrival in the country and population by place of residence one year ago.

Emigration estimates are based on three data sources – emigration stated in the Register of Natural persons, emigration indicated in the results of the Population and Housing Census 2011 and estimated emigrants based on both administrative registers and Population and Housing Census 2011. With the help of this information the number of emigrants is acquired on the time period 2000–2010. Number of emigrants has been recalculated in a way to ensure that several demographic indicators in gender and birth year groups are harmonised in time, i.e.:

- Birth number from 2000 till 2010;
- Death number from 2000 till 2010;
- Immigration from 2000 till 2010;
- Population number at the beginning of 2000 and 2011.

As Register of Natural persons and Population and Housing Census 2011 provides information on emigrant’s gender, birth year, emigration year and emigration times, while administrative sources on estimated emigration give information only on birth year and gender, thus to estimate the total number of emigrants in breakdown by year a "raking ratio" calibration method is used [1].

The calibration method is based on the data regarding each emigrant’s emigration year in Register of Natural persons and Population and Housing Census 2011 as well as total all emigrants’ number of emigrations during the time period from 2000 till 2010 that have been calculated with the help of equation:

P²⁰¹¹ = P²⁰⁰⁰ + DZ^2000-2010 - M^2000-2010 + IM^2000-2010 - EM^2000-2010, where

P²⁰¹¹ – population number at the beginning of 2011;
P²⁰⁰⁰ –  population number at the beginning of 2000;
DZ^2000-2010 – birth number from 2000 till 2010;
M^2000-2010 – death number from 2000 till 2010;
IM^2000-2010 – immigration from 2000 till 2010;
EM^2000-2010 – emigration from 2000 till 2010.

As a result, weights in gender and birth year groups were calculated for each emigrant stated in the Register of Natural persons and Population and Housing Census 2011, and they were used to estimate the total emigrant number in breakdown by year.

It should be considered that in cases, when emigrant was stated in both Register of Natural persons and Population and Housing Census 2011, a decision was made to use the emigration year mentioned in the Population and Housing Census 2011.

References: [1] Jean-Claude Deville and Carl-Erik Sarndal. (1992) Calibration Estimators in Survey Sampling. Journal of the American Statistical Association, 87 (418): 376–382.