House prices are some of the few financial goods that are modelled only to a limited extent in microeconomic systems, despite the heterogeneity of real estate. Usually, no distinction is made between the type of property. This paper addresses this gap by examining property prices of privately-owned apartments in the 17th district of Vienna. First, price indices are created from land register sales data using various models to determine the current price level. Factors influencing real estate prices are identified and Machine Learning models are used to forecast the price development in the area of interest. Finally, the broader implications at the micro- and macroeconomic scale are discussed.
Table of Contents
1 Introduction
2 Methodology for modelling current and future property prices in the 17th district of Vienna
2.1 Methods of calculating the current price level of privately-owned apartments
2.1.1 Price level determination through plain time series (TS)
2.1.2 Price level determination through random forest algorithm (RF)
2.1.3 Price level determination through hedonic models
2.2 Approach to forecasting price level developments
2.2.1 Random forest
2.2.2 eXtreme Gradient Boosting (XGBoost)
2.2.3 Support vector machines/regression (SVM)
2.2.4 Prophet (with XGBoost errors)
2.2.5 Auto regressive integrated moving average (ARIMA) (with XGBoost errors)
2.2.6 Linear Model (only in TS)
3 Determining influencing factors
3.1 External/macroeconomic influencing factors
3.1.1 Supply and demand
3.1.1.1 The population development
3.1.1.2 Purchasing power
3.1.1.3 Rate of inflation
3.1.1.4 Interest rates
3.1.1.5 Improvement of infrastructure
3.1.2 Transaction volume
3.2 Internal influencing factors
3.2.1 Location factors
3.2.2 Useable area
3.2.3 Sum of loggia areas
3.2.4 Noise
3.3 Narratives and experience
4 Results
4.1 Results of modelling the current price level of privately owned apartments in the 17th district of Vienna
4.1.1 Plain time series analysis
4.1.2 Random forest algorithm
4.1.3 Hedonic model
4.1.4 A data analysis of estimates from Austrian institutions
4.2 Forecasting results
4.2.1 Price level forecast through regular time series
4.2.2 Price level forecast through random forest regression
5 Analysis
5.1 Social and economic implications
5.1.1 Effects on the individual
5.1.2 Effects on the national economy
6 Conclusion
7 Evaluation
7.1 Evaluation of the study
7.2 Suggestions for further research questions
Research Objective And Core Topics
This paper aims to investigate the factors influencing current real estate price levels in the 17th district of Vienna (Hernals) and to develop robust forecast models for future price developments, addressing the scarcity of granular microeconomic modeling for property assets.
- Application of Machine Learning models (Random Forest, XGBoost, SVM) for price forecasting
- Economic analysis of macroeconomic factors like inflation, interest rates, and demographic shifts
- Evaluation of internal property characteristics (e.g., location, usable area, loggias) affecting value
- Assessment of behavioral economic factors (narratives) and their impact on market transparency
- Comparative analysis of institutionally aggregated price indices versus model-based calculations
Excerpt From The Book
3.3 Narratives and experience
A third, significant influencing factor which has so far remained largely unconsidered among economists, is the importance of general public opinion or the narrative that drives certain economic developments.
To quote Hume in a context of “cultural entrepreneurship” (Mokyr, 2017), this implies that, “What depends on a few persons is, in great measure, to be ascribed to chance, or secret and unknown causes; what arises from a great number may often be accounted for by determinate and known causes” (Hume, 1742/2016).
On the one hand, Shiller postulates that it is important to recognise economic narratives in order to counteract them, but also to be prepared for possible extraordinary changes in markets. On the other hand, one becomes painfully aware that there is as yet no exact science or method of classifying these narratives (Shiller, 2019).
It should therefore be made clear at this point that this theory cannot be substantiated with scientific evidence at present.
The seemingly most effective narrative at the moment is that real estate is the safest investment, which never loses its value and so demand continues to grow due to the increase in population. This is often cultivated by the media, although they have comparatively little insight into the real estate market due to a lack of data. As a result, a part of the population sees property, exaggeratedly formulated, as a “religious duty” and thus irrational exuberance sets in (Shiller, 2000). On the other hand, the "squeezing out" of citizens from the market by large investment firms is leading to fewer and fewer people buying property and instead increasingly renting. Particularly due to the fast-paced, changing world of work, it is also no longer of great importance for many families
Summary of Chapters
1 Introduction: Discusses the basic need for housing, the rising trend of urbanization in Austria, and sets the research context for examining the 17th district of Vienna.
2 Methodology for modelling current and future property prices in the 17th district of Vienna: Details the quantitative and Machine Learning approaches, including time series analysis, random forests, hedonic models, and various regression techniques used for forecasting.
3 Determining influencing factors: Analyzes both macroeconomic drivers like supply, demand, inflation, and interest rates, and micro-level property features, alongside the role of market narratives.
4 Results: Presents the findings from the different models, contrasting empirical data with institutional estimates and outcomes of the forecast models up to 2025.
5 Analysis: Explores the socioeconomic impacts of the identified price trends, specifically focusing on individual affordability, wealth distribution, and broader national economic implications.
6 Conclusion: Synthesizes the results, summarizing expected price growth and discussing the inherent limitations concerning real-world interrelations that models cannot fully capture.
7 Evaluation: Reflects on the study's scope, validity of methods, data transparency issues, and potential avenues for future scientific inquiry.
Keywords
Real Estate, Vienna 17th District, Hernals, Property Price Forecasting, Machine Learning, Random Forest, Hedonic Models, Supply and Demand, Macroeconomics, Property Valuation, Market Transparency, Wealth Inequality, Housing Market, Investment Narrative, Time Series Analysis
Frequently Asked Questions
What is the primary focus of this research?
The paper examines the real estate market in Hernals (17th district of Vienna) to identify factors driving current price levels and to forecast future developments using various statistical and Machine Learning models.
What are the central thematic areas covered?
The work covers theoretical price modeling, the influence of macroeconomic factors (interest rates, inflation), property-specific internal characteristics, and the impact of social narratives on the housing market.
What is the core research question being addressed?
The study primarily asks how real estate prices will develop in the near future and what the resulting social and economic consequences are for individual consumers and the national economy.
Which scientific methodology is utilized in this study?
The author uses a variety of methods including time series analyses, Random Forest regression, XGBoost, and hedonic pricing models (GAM) to process and project transaction data.
What constitutes the main body of the work?
The main body comprises a methodological breakdown of current value determination and future price forecasting, a detailed analysis of influencing factors, and an evaluation of the societal and economic implications of these trends.
Which keywords best characterize this work?
Key terms include Real Estate, Vienna Hernals, Property Forecasting, Machine Learning, Random Forest, Hedonic Models, and Macroeconomic Influencing Factors.
Why was the 17th district of Vienna chosen for the study?
The 17th district (Hernals) was selected because it is the author's home district, providing a localized, manageable area to investigate that reflects broader trends within the Viennese residential market.
What role do "narratives" play in this specific research?
The research explores the behavioral economic aspect of how general public opinion and the media-driven narrative—that real estate is an indestructible asset—significantly impact market activity and price formation.
- Quote paper
- Yan S. Matschi (Author), 2023, An examination of the real estate price development in the 17th district of Vienna, Munich, GRIN Verlag, https://www.grin.com/document/1460000
