Accuracy and reliability

Overall accuracy

The estimated number of sales is slightly underestimated, as some transactions are only registered more than a year after the transaction and are therefore not included in the statistics. In addition, pile sales are not included. It is estimated at a maximum of 200 per year.

There is no reason to believe that the average price is neither overvalued or undervalued in the final figures, as the registered prices are judged to be fair. However, preliminary figures for the average prices are underestimated because no correction is made for the more expensive properties to be included in the database later than the cheap properties.

As the appraisal values have been maintained since 2011, the purchase sum in percent of taxable value (spar-value) is higher than it would otherwise be if the appraisals had been continuously updated every year.

There is no knowledge of systematic over- or underestimation of the final figures for the price index. For the preliminary figures for price development, bias and deviation have been calculated in: analysis: The reliability of the preliminary figures for house price development for the first figures for the price development in relation to the final figures in 2019. The analysis shows, for example, that the first estimate for the price development for one-family houses in a given month in 2019 compared to the same month the year before differs by an average of 0.7 percentage points from the final estimate. The corresponding figure for owner-occupied flats is 0.8 percentage points. The analysis also shows that there is no significant systematic bias in the estimates. Overall, there is a slight negative bias of 0.1 percentage points for both owner-occupied flats and one-family houses. In the StatBank (tables VERSEJ14, VERSEJ55, [VERSEJ12](https: //www.Statbank.dk/VERSEJ12) ​​and VERSEJ66) there are versiontables for the price index, which show the preliminary and final figures, so it is possible to assess how close the preliminary figures are on the final figures.

Sampling error

This statistic is a total statement of all registered real estate transactions based on administrative register information.

Non-sampling error

Two identical properties with the same quality and location can be traded at different prices at a given time, as there is no perfect and transparent market. For example, a property may obtain a higher price if the seller has been waiting for the "right buyer" and there is always an element of random in the pricing. The public appraisal value is also subject to uncertainty, so that two identical properties do not necessarily have the same appraisal value. The spread or uncertainty of the appraisal values together with the uncertainty of the traded property prices for the period, contributes to the uncertainty of the spar-value and thus to the uncertainty of Statistics Denmark's property price index.

The uncertainty in the property price is higher when the geographical area or the observation period is reduced. For example, the uncertainty in the property price is higher for monthly figures in the Region Nord Jylland than for annually figures nationwide. It is also estimated that the uncertainty in the price index for commercial properties is higher than for dwellings, as relatively few commercial properties are sold compared to dwellings and at the same time commercial properties are often more heterogeneous than dwellings, which makes it difficult to quality adjust the price. The measurement uncertainty of quality correction is described in the analyze Usikkerhed ved opgørelse af udviklingen i boligpriser, only available in Danish.

The measurement uncertainty of quality correction is described in the analyze Usikkerhed ved opgørelse af udviklingen i boligpriser, only available in Danish.

Data is received from an administrative register, where error registrations may occur. However, it is not considered that incorrect registration of the selling price and incorrect appraisal values have a significant effect on the published figures, as the ratio of sales price and appraisal value for each individual transaction (spar-value) is checked on an ongoing basis. Incorrect appraisal values are considered to be of a greater extent for commercial properties than for dwellings, as relatively few commercial properties are sold and they are at the same time more heterogeneous than dwellings.

The electronic registration has made it more difficult to identify so-called multiple sales of owner-occupied flats. The phenomenon is particularly pronounced for parent properties, which consist of several rental properties, but also exist for building sites. Multiple sales include approximately 100 transactions per year and have not been included in the statistics since 2011.

There are problems with correct registration of partial sales (sale of a subset of a property), which particularly affects agricultural properties. The sale of agricultural land from a larger agricultural property is an example of a partial sale. The problem arises because the partial sale does not have a corresponding appraisal value. Thus, the sales price relates to a partial sale, while the appraisal value relates to the entire property. This is a problem because the ratio between the sales price and appraisal value (spar-value) is used to calculate the price index. The partial sales that are incorrectly registered as full sales in the land registry system are not considered to be a source of great uncertainty in the price development, as the ratio between the sales price and appraisal value (spar-value) is calculated for each individual property and it is checked whether it is in the range of 0,4 - 3. Following the transition to electronic registration on 8. September 2009, fewer transactions has been registered as partial sales. The reason for this is unknown. The variable for partial sale is described in more detail on Statistics Denmark's website for high-quality documentation under property sales.

The codes for type of transfer are defined by the Danish Tax Authorities and are included in the data that the land registration system send to the Danish Tax Authorities for use in calculate the public appraisal values. It does not appear directly from the guidelines from the Danish Tax Authorities how parental purchase that are sold to their child must be registered, but it is the experience of the land registry court that most people choose "intra-family sale", when it comes to sales from parents to children.

It is not considered that failure to report to the electronic land registry system is a significant source of uncertainty, as it is mandatory to report sales information in connection with the sale of real property in Denmark. It may happen that a real property transaction is registered very late after it has been sold and that it therefore does not have time to be included in the final figures, which is published approximately thirteen months after the transaction date. These are approximately 100 transactions per year, which is less than 1 per mille of all sales.

Analyzes shows that around 70 per cent of all transactions are included in the first publication of the monthly figures. For the second publication it is around 85 per cent.

The freedom code for owner-occupied flats is often missing, and in those cases it is assumed that the dwelling can be freely traded. The freedom code indicates whether the owner-occupied flat is free from the obligations of the tenancy legislation, which i.a. protects previous tenants against termination.

The calculation of the price index for properties is complicated by the fact that different properties with different characteristics are sold from quarter to quarter (period to period), so it is not possible to easily compare prices in two quarters (periods). To make a comparison anyway, one has to quality correct the property prices. This can be done in several ways which are mentioned in the analysis Sammenligning af boligprisstatistikker i Danmark, only available in Danish. No matter what method is chosen, a quality correction of property prices will never be completely perfect. This applies to all methods, including Statistics Denmark's SPAR-method and the housing organizations calculation of the price per square meter. The SPAR-method (Sales Price Appraisal Ratio method) has been chosen as it can correct for more than location and dwelling size. The quality correction of the method is made by measuring the ratio of sales price and appraisal value (spar-value). The development in property prices is determined as the development in the average spar-value. The SPAR-method is also used by the statistical offices in Sweden and the Netherlands. Statistics Denmark does not continuously make statistics with other methods than the officially chosen one, but occasionally other methods are tested as a check. Statistics Denmark's official method is described in more detail in the note: Beregning af prisindeks for ejendomssalg, only available in Danish. The described quality correction, where the ratio of sales price and appraisal value is calculated assumes that the appraisal value is a sufficiently fair bid on the sales price at the time of valuation, sufficient to be able to use the SPAR-method.

It is well known that the public appraisal value is subject to uncertainty, so it can not be assumed that two identical properties with a uniform location are always valued at the same price. The premise is instead that the appraisal value is, on average, sufficiently accurate. For the purposes of the property price index, the country is divided into geographical areas, and it is assumed that any difference between the sales price and the appraisal values is systematic. In that case the systematic is eliminated or at least significantly reduced when calculating the relative development in the ratio of sales price and appraisal value.

When calculating the price index, "extreme observations" are removed. The significance of the limits for the individual spar-value is at the moment examined in more detail.

Statistics Denmark periodically examines whether the price index is affected by whether transactions weights or stock weights are used, see Model group paper about house prices, only available in Danish.

Quality management

Statistics Denmark follows the recommendations on organisation and management of quality given in the Code of Practice for European Statistics (CoP) and the implementation guidelines given in the Quality Assurance Framework of the European Statistical System (QAF). A Working Group on Quality and a central quality assurance function have been established to continuously carry through control of products and processes.

Quality assurance

Statistics Denmark follows the principles in the Code of Practice for European Statistics (CoP) and uses the Quality Assurance Framework of the European Statistical System (QAF) for the implementation of the principles. This involves continuous decentralized and central control of products and processes based on documentation following international standards. The central quality assurance function reports to the Working Group on Quality. Reports include suggestions for improvement that are assessed, decided and subsequently implemented.

Quality assessment

The data on which the calculation of the statistics for sales of real properties is based are generally of a high quality, as they come from the mandatory electronic real property registration system, which is the most comprehensive register of real property transactions in Denmark. To ensure high quality of the statistics, all calculations are performed mechanically in a dedicated production system and various forms of validity checks of both input data and results are performed on an ongoing basis.

The calculation of the price index for properties is complicated by the fact that different properties with different characteristics are sold from quarter to quarter (period to period), so it is not possible to easily compare prices in two quarters (periods). To make a comparison anyway, one has to quality correct the property prices. This can be done in several ways which are mentioned in the analysis Sammenligning af boligprisstatistikker i Danmark, only available in Danish. No matter what method is chosen, a quality correction of property prices will never be completely perfect. This applies to all methods, including Statistics Denmark's SPAR-method and the housing organizations calculation of the price per square meter. The SPAR-method (Sales Price Appraisal Ratio method) has been chosen as it can correct for more than location and dwelling size. The quality correction of the method is made by measuring the ratio of sales price and appraisal value (spar-value). The development in property prices is determined as the development in the average spar-value. The SPAR-method is also used by the statistical offices in Sweden and the Netherlands. Statistics Denmark does not continuously make statistics with other methods than the officially chosen one, but occasionally other methods are tested as a check. Statistics Denmark's official method is described in more detail in the note: Beregning af prisindeks for ejendomssalg, only available in Danish. The described quality correction, where the ratio of sales price and appraisal value is calculated assumes that the appraisal value is a sufficiently fair bid on the sales price at the time of valuation, sufficient to be able to use the SPAR-method.

Furthermore, a bias adjustment is made in the preliminary price indices to correct for the observed skewness in the registration and thus the reporting pattern. Due to the desire for statistics shortly after the end of the reference period, a number of preliminary versions of the statistics are compiled before all observations for a given period are available and thus it is possible to compile more accurate statistics.

Data revision - policy

Statistics Denmark revises published figures in accordance with the Revision Policy for Statistics Denmark. The common procedures and principles of the Revision Policy are for some statistics supplemented by a specific revision practice.

Data revision practice

All figures are revised 11 times, both monthly, quarterly and yearly figures. The calculation is made on the basis of the transactions which have been registered in the electronic land registration system at the time of calculation. The preliminary figures are calculated based on an incomplete set of data since not all transactions have been registered at the time of calculation.

The preliminary figures for the price development are continuously revised for two reasons: (1) The figures are calculated on the basis of an incomplete set of data, since not all transactions have been registered at the time of calculation. (2) Changes in the appraisal values as the Tax Authorities makes a new appraisal value every second year. The appraisal values is usually included in the calculation from and including the December publication. This means that for the period January - November, the first publications are based on the old valuation (appraisal) year, but are subsequently revised to the new valuation. The current appraisal value will be the most accurate to use, as it takes into account recent changes in the general price level and in the building stock.

Analysis made by Statistics Denmark shows that the transactions of properties is typically registered into the electronic land registration system close to the acquisition date, which is the time when the new owners gets the key to their new property. The acquisition date may be relatively long time after the transaction date. The analysis also shows that the period between the transaction-date and the acquisition date is longer for more expensive properties than for cheaper properties. At the same time analysis also shows that expensive properties is more likely to have a relatively lower appraisal than cheaper properties. That is, the appraisals for expensive properties is typically less than the sales price while it is opposite for the cheaper properties where the appraisals typically is higher than the sales price. The preliminary price indices are calculated in a way that adjust for these circumstances in order to keep the bias and deviation as low as possible. The method is described in detail: here.

The preliminary figures for average prices are underestimated because no correction is made for that the more expensive properties are included later in the database than the cheap properties (timing issues).

For the preliminary figures for the price development, the analyze Pålideligheden af de foreløbige tal for boligprisudviklingen, only available in Danish, calculate bias and deviation for the first figures for the price development in relation to the final figures in 2019. The analysis shows, for example, that the first estimate for the price development for one-family houses in a given month in 2019 compared to the same month the previous year deviates on average by 0.7 percentage points from the final estimate. The corresponding figure for owner-occupied flats is 0.8 percentage points. The analysis also shows that there is no significant systematic bias in the estimates. Overall, there is a slight negative bias of 0.1 percentage points for both one-family houses and owner-occupied flats.

In the Statbank (tables VERSEJ14, VERSEJ55, [VERSEJ12](https: //www.Statbank.dk/VERSEJ12) ​​and VERSEJ66), there are versiontables for the price index for dwellings, which show the preliminary and final figures, so that it is possible to assess how close the preliminary figures are to the final figures.