scholarly journals Assessment of Predictor Importance with the Example of the Real Estate Market

2016 ◽  
Vol 16 (2) ◽  
pp. 29-39
Author(s):  
Mariusz Kubus

Abstract Regression methods can be used for the valuation of real estate in the comparative approach. However, one of the problems of predictive modelling is the presence of redundant or irrelevant variables in data. Such variables can decrease the stability of models, and they can even reduce prediction accuracy. The choice of real estate’s features is largely determined by an appraiser, who is guided by his/her experience. Still, the use of statistical methods of a feature selection can lead to a more accurate valuation model. In the paper we apply regularized linear regression which belongs to embedded methods of a feature selection. For the considered data set of real estate land designated for single-family housing we obtained a model, which led to a more accurate valuation than some other popular linear models applied with or without a feature selection. To assess the model’s quality we used the leave-one-out cross-validation.

2021 ◽  
pp. 52-66
Author(s):  
Huang-Mei He ◽  
Yi Chen ◽  
Jia-Ying Xiao ◽  
Xue-Qing Chen ◽  
Zne-Jung Lee

China has carried out a large number of real estate market reforms that change the real estate market demand considerably. At the same time, the real estate price has soared in some cities and has surpassed the spending power of many ordinary people. As the real estate price has received widespread attention from society, it is important to understand what factors affect the real estate price. Therefore, we propose a data analysis method for finding out the influencing factors of real estate prices. The method performs data cleaning and conversion on the used data first. To discretize the real estate price, we use the mean ± standard deviation (SD), mean ± 0.5 SD, and mean ± 2 SD of the price and divide it into three categories as the output variable. Then, we establish the decision tree and random forest model for six different situations for comparison. When the data set is divided into training data (70%) and testing data (30%), it has the highest testing accuracy. In addition, by observing the importance of each input variable, it is found that the main influencing factors of real estate price are cost, interior decoration, location, and status. The results suggest that both the real estate industry and buyers should pay attention to these factors to adjust or purchase real estate.


2019 ◽  
Vol 12 (2) ◽  
pp. 250-266 ◽  
Author(s):  
Jacek Batóg ◽  
Iwona Foryś ◽  
Jan Konowalczuk

Purpose This study aims to present a problem of noise compensation related to the localization of single-family houses in the restricted use areas (RUA) created around airports. The authors presented different methodological approaches to the valuation of such amends and characterized distinct solutions applied in that area in chosen countries and cities. Design/methodology/approach To estimate the level of those indemnities, linear models including spatial and generalized linear models, were applied. The set of explanatory variables contains quantitative and qualitative attributes of real estates. The influence of outliers indicated by means of cluster analysis on the received results were also considered. Findings The results show that after accounting for noise level and house characteristics, houses in noisier areas were sold for less than houses subjected to less noise. Unit prices of houses located outside the RUA were on average 17.05 per cent higher than the prices observed in zone with noise level for the daytime of 60 dB and 8.95 per cent in zone characterized by noise level for the daytime of 55 dB. Practical implications Received results can be compared with results obtained by other authors, but its most important application is possibility of use the proposed methodology by judicial appraisers to assess the proper level of noise compensation for home owners or tenants. Originality/value An identification of real estate market heterogeneity and its considering in estimation of compensation related to airport noise evidence some novelty of the research.


2014 ◽  
Vol 22 (2) ◽  
pp. 5-12
Author(s):  
Justyna Tanaś

Abstract The market for land for single-family housing is one of the most important segments of the real estate market. Over the last several years, we have witnessed the development of this market in the suburban areas of large agglomerations. This is connected with the intensified migration of people from the centers of big cities, and with the development of housing in suburban zones. The aim of this paper is to present the structure of the territorial origin of buyers purchasing plots designated for single-family housing in Poznań and its suburbs (with division into buyers derived from the city of Poznań, from the Poznań poviat (district), the Wielkopolska voivodeship (province), the rest of the country, and from abroad). The analysis covers the years 1995-2010.


Author(s):  
Anna Przewiezlikowska

The aim of this article is the comparative description of two real estate markets based on the procedures for real property valuation. The study concerned only the land, which was undeveloped, intended for single-family housing in two communes located in the district of Krakow and three communes from the district of Kielce. The analyses were performed at four-year intervals and the comparison of the real estate markets was conducted. The first part contains the description of the areas covered by the research studies and the analyses of the real estate market and market trends. The next stage includes the descriptions of the two test real properties which are the subject of valuation and the fundamental comparative criterion. Then, the algorithms and methods of the calculations are presented. The practical part contains the description of individual markets, the implementation of the analyses and calculations, the comparison of the study areas and conclusions. The comparative analysis of the performed simulations of valuations was carried out first and then followed by a collective summary of descriptive statistics of all the real estate bases and the comparative description of the structures of the databases showing meaningful differences between Krakow and Kielce region.


2019 ◽  
Vol 27 (4) ◽  
pp. 15-26
Author(s):  
Krzysztof Dmytrów ◽  
Sebastian Gnat

Abstract Property valuation in the comparative approach requires the determination of the impact of market characteristics on the formation of prices on the local real estate market. Valuers have a variety of methods for determining weights. Some of them require the collection of a sufficiently large database of information on transactions. However, this is not always possible. In the absence of sufficient data, alternative approaches, including an expert approach, may be used. The goal of the article is the proposal of an expert approach at the stage of assessing the influence of attributes on the value of the real estate. The AHP (Analytical Hierarchy Process) method will be used. On its basis, pairwise comparisons of the importance of attributes will be done by experts (valuers). By means of the AHP method, the weights of each attribute will be obtained and, subsequently, the influence of each attribute on the real estate value will be assessed. Research will be done on the basis of 318 real estates in Szczecin.


2016 ◽  
Vol 24 (4) ◽  
pp. 70-78 ◽  
Author(s):  
Alina Źróbek-Różańska

Abstract Predicting demand on the residential real estate market and the behavior of the purchasers requires a wide knowledge of both the economic mechanisms and psychology of decision-making. Decisions on the real estate market are often made by people without professional skills, and using simplified strategies. However, the decision-making process, on top of its heuristic nature, is dynamic and changing. As a result, a discrepancy in the preferred characteristics of planned and actually bought real estate can be observed. Such a discrepancy can be explained with the occurrence of the compensation process. The aim of this article was to recognize and describe the compensation process on the example of the suburban residential real estate market. The aim was achieved by analyzing the preferences of potential buyers in terms of particular characteristics of the location of suburban plots destined for single-family housing (respondents divided into age groups: 25 and 26-40), analyzing the real settlement trend in the suburban zone (the result of actual transactions) and comparing the results, including compensation.


2016 ◽  
Vol 24 (3) ◽  
pp. 40-51
Author(s):  
Małgorzata Renigier-Biłozor ◽  
Andrzej Biłozor

AbstractPreliminary data analyses in decision-making systems and procedures are very important for numerous reasons, in particular because the accumulation and analysis of large data sets is costly and time-consuming. The effective use of decision support systems, including on the real estate market, requires the elimination of noise. The authors have proposed to eliminate redundant data with the use of the modified method for evaluating the capacity of the data set, which is applied in the process of classifying the condition of real estate markets. The proposed procedure (subsystem) is an attempt to improve the effectiveness of analyses relating to the development of methods for rating real estate markets. The proposed solutions will be simulated on the example of leading real estate markets in Poland and Italy.


2019 ◽  
Vol 27 (2) ◽  
pp. 33-43
Author(s):  
Tomasz Adamczyk ◽  
Agnieszka Bieda ◽  
Piotr Parzych

Abstract The complexity of multi-component real properties results from the possibility of identifying various components in legal, physical or functional terms. The possibility of distinguishing various functional elements of real properties, combined with the specificity resulting from their market properties, is problematic when applying the comparative approach to real estate valuation. In this case, the valuation procedure can be implemented using statistical models: the parametric model or the conditional one. This research paper demonstrates the construction of the parametric and conditional models taking into account the geometric and pricing attributes of multi-component real estate. The authors paid attention to adjusting the models to the available market data. They also specified the conditions for the use of statistical models in the real estate valuation process. Based on the analytical and accounting considerations, the estimation criteria for the parametric model and the conditional model were defined, which allow the correct application of these models at the stages of the real estate market analysis and the real estate valuation process.


2020 ◽  
Vol 6 (2) ◽  
pp. 82-87
Author(s):  
Victoria V. Skorinova ◽  
Elena I. Lobanova

The article is a continuation of the study of price zoning. Earlier in the article "Research of approaches to price zoning of localities" theoretical apparatus of the study was studied, the procedure for conducting price zoning was determined, and the main approaches to price zoning of localities were analyzed. The purpose of this study is to analyze the price zoning of localities as a cost factor in real estate valuation. The article summarizes the results of analysis of the Russian real estate market. The views of many researchers on the terms "price zoning" and "valuation zoning"are analyzed. In the field of price zoning, a statistical analysis of expert opinions of professional participants in the real estate market has carried out, in addition to expert assessments, the regulatory framework for valuation activities were analyzed. The relevance of the study of price zoning of the territory of localities in the work of the appraiser with a comparative approach to assessing the value of real estate is shown.


Sign in / Sign up

Export Citation Format

Share Document