Latvia - House Price Index

Latvia: House Price Index

Mnemonic HPI.ILVA
Unit Index 2015=100, NSA
Adjustments Not Seasonally Adjusted
Quarterly 4.31 %
Data 2018 Q4 133.87
2018 Q3 128.34

Series Information

Source Central Statistical Bureau of Latvia
Release House Price Index
Frequency Quarterly
Start Date 3/31/2006
End Date 12/31/2018

Latvia: Real Estate

Reference Last Previous Units Frequency
Building Permits 2019 Q1 1,022 1,001 #, NSA Quarterly
Non-residential Building Permits 2019 Q1 361 286 #, NSA Quarterly
Residential Building Completions 2019 Q1 115.5 149.6 Ths. m², NSA Quarterly
Residential Building Permits 2019 Q1 661 715 #, NSA Quarterly
House Price Index 2018 Q4 133.87 128.34 Index 2015=100, NSA Quarterly
House Price Index for Existing Homes 2018 Q4 135.7 130.14 Index 2015=100, NSA Quarterly
House Price Index for New Homes 2018 Q4 123.61 118.27 Index 2015=100, NSA Quarterly

Release Information

For Latvia, the House Price Index (HPI) reflects changes in dwelling prices in a specified period of time. New dwellings, existing dwellings, total. Quarterly from 2006.

House Price Index (HPI) - a quarterly indicator that measures the price changes of dwellings that households purchase on the market. The HPI covers all transactions of dwellings regardless of the final use of the dwelling. This index covers both the transactions that are new to the household sector (purchased from legal entity, municipality, government), and also trade between households. Prices include land value.

New dwellings – new flats and new single-dwelling houses.

New flats – prior unoccupied flats in multi-dwelling houses that are sold to individuals within 3 years after commissioning.

New single-dwelling houses – prior unoccupied, fully completed buildings fit for human-occupancy that are sold to individuals.

Existing dwellings – existing flats and single-dwelling houses.

Existing flats – prior occupied (by legal entity, municipality, government, other individuals) apartments that are sold to individuals.

Existing single-dwelling houses – prior occupied (by legal entity, municipality, government, other individuals) houses that are sold to individuals.

The source writes:


In compliance with the, Draft of the Regulation laying down detailed rules for the implementation of the Council Regulation (EC) No 2494/95 concerning harmonised indices of consumer prices, as regards establishing owner-occupied housing price indices the HPI is published at two levels of detail with the following structure:
1. H.1. Dwellings (total);
2. H.1.1. New dwellings;
3. H.1.2. Existing dwellings.

Collection of data

  • Survey method and data sources

Main data source used to obtain information is the State Unified Computerised Land Register. The index is calculated basing on the transaction prices registered in the Land Register by individuals during the reference period, whereas weights reflect structure of the dwelling purchases during the previous calendar year. Weights are revised every year and recalculated into prices of the 4th quarter of the previous year.

  • Target population

Coverage of population:
HPI covers all transactions in which individual(s) is indicated as buyer, regardless of legal status of the seller(s). HPI includes also dwelling purchases made within the economic area of Latvia by individuals - non-residents.

HPI excludes transactions between legal entities, municipalities and government, as well as deals engaging legal entity as a buyer and individual as a seller.

Index does not include commercial areas, outhouses, saunas, garages, cellars, gazebos and other structures unfit for human occupancy.

Geographical coverage:
Whole territory of the Republic of Latvia.


House Price Index is a Laspeyres-type quality-adjusted price index. For the index calculation all transactions are stratified into the following groups (strata): new flats, new single-dwelling houses (till 2008, including), existing flats in Riga, existing flats in the rest of the country, existing single-dwelling houses.

Group index is calculated as average geometric transaction sum ratio between reference and base periods.
In order to adjust structural (qualitative) dwelling transaction differences between the reference and base periods, a hedonic regression method is used within each group. The method based on the most significant price determinants is used for dwelling quality value calculations.

Ln transaction sum = β0 + β1χ1 + β2χ2 +…+ βiχi i = 1,….., n,


Ln transaction price – natural logarithm of transaction price;
xi – price factors;
βi – regression coefficients (factor impacts);
β0 – constant.
Factors having the greatest impact on dwelling price and taken into account in the HPI calculations are geographical location of the property, property size (area) and availability of amenities.
For the calculation of higher level price indices and the overall HPI a Laspeyres-type formula is used that expresses the weighted arithmetic mean value of the lower level price indices:

the overall HPI in the period T, compared to period 0 (base period);

share of the transaction price in stratum j in a base period;

price index in stratum j in a period T, compared to period 0 (base period).

  • Base period

Base period used in the price calculations is the 4th quarter of the previous year. To calculate price changes during a longer time span, price indices of each year are “chained” in one dynamic series having the same comparison period.

Comparison periods of the HPI are years 2010 and 2015, expressed with 100 index points (2010=100 and 2015=100). Moody's Analytics collects data calculated with respect to both the base years. Price indices are used to calculate price changes over the previous quarter, corresponding quarter of the previous year or any other time period.

If information used for the calculations is revised or corrections are received, the published data may be revised and changes are considered in the running publication. Mistakes in calculations are corrected immediately after their detection.

Moody's Analytics collects data for both the base years (2010=100 and 2015=100).