© Gabriel M. Ahlfeldt, Tobias Seidel
Version 0.91, 2024
This toolkit covers the algorithm by Ahlfeldt, Heblich, Seidel (2023) to generate micro-geographic property price and rent indices for arbitrary spatial units based on transaction data. It includes a replicable analysis based on artificial data for illustration.
This toolkit also includes the indices created by Ahlfeldt, Heblich, Seidel (2023) for Germany. We keep updating these indices as time goes by. These micro-geographic property price indices are based on real micro-data from Scout24. They can be found in the following sub-folder: APPLICATIONS/DATA/OUTPUT. For replication, the underlying microdata must be accessed from the Forschungsdatenzentrum at RWI-Essen.
We provide an interactive webtool to illustrate the latest edition of the post-code level price and rent indices for Germany here.
When using the toolkit in your work, please cite Ahlfeldt, Heblich, Seidel (2023): Micro-geographic property price and rent indices. Regional Science and Urban Economics, 98. https://doi.org/10.1016/j.regsciurbeco.2022.103836
You can download the toolkit and use it as a starting point for your own analysis. The algorithm requires two input files containing your observed market transactions and the centroids of the spatial units for which you wish to generate the index. These are stored in the folder DATA/INPUT. You can replace these with your transaction data sets and your coordinates. Please ensure that the files you replace them with are in exactly the same format; i.e. you should use exactly the same variable names. All coordinates need to be in projected units of the same dimension. You may use a different ID variable name in the centroids.dta file as long as you update that variable name in the DO/MASTER.do do file.
Please consider the README files in the respective folders for further detail.
| Directory | Sub-folder | Description | Additional Information |
|---|---|---|---|
ADOS |
Folder containing ado files called by the algorithm | - | |
APPLICATIONS |
DATA/OUTPUT |
Folder containing indices generated by Ahlfeldt, Heblich, Seidel (2023) and shapefiles for illustration | This folder contains indices computed based on real data |
APPLICATIONS |
DATA/OUTPUT/2022 |
2022 edition of the postcode indices including years 2007-2021 | This folder contains indices computed based on real data |
APPLICATIONS |
DATA/OUTPUT/2024 |
2024 edition of the postcode indices including years 2007-2023 | This folder contains indices computed based on real data |
APPLICATIONS |
DATA/XWALKS |
A subfolder containing crosswalks from postcodes to counties and local labour markets, as well as postcode population that may be used for weighted aggregation | - |
DATA |
Folder containing input data and output data for the replicable analysis to trial the algorithm | This folder contains artificial data | |
DO |
Folder containing do files for the replicable analysis based on artificial data | - | |
MAPS |
Folder containing maps generated by the replicable analysis based on artificial data | ||
MAPS |
PURCH-PLZ-2024 |
Mapped 2024 edition of postcode purchase price index | |
MAPS |
RENT-PLZ-2024 |
Mapped 2024 edition of postcode rent price index | |
SHAPES |
Folder containing the shapefile of hexagons used in the replicable analysis based on artificial data | - |
Ahlfeldt, G. M., Heblich, S., Seidel, T. (2023): Micro-geographic property price and rent indices. Regional Science and Urban Economics, 98. https://doi.org/10.1016/j.regsciurbeco.2022.103836
Alexander Hansen has created an R version of the toolkit which is available here: https://github.com/hvervetid/housepriceindex
0.91: Added an additional check for sufficient variation in covariates in locally weighted regressions
13/11/2024: Added German postcode rent and price indices from 2007-2023 22/11/2024: Added German municipality (Gemeinde) rent and price indices from 2007-2023 13/12/2024: Added German local labour market (Arbeitsmarktregion) rent and price indices from 2007-2023
Acknowledgement:
I thank Alexander Hansen for spotting a bug in an earlier version of the code. I thank Vincent Heddesheimer for helping with the creation of the 2024 municipality (Gemeinde) index.