关于巴西房地产估价的一点说明

Q4 Economics, Econometrics and Finance

Revista Brasileira de Economia Pub Date : 2021-07-12 DOI:10.5935/0034-7140.20210003

Thiago Marzagão, R. Ferreira, Leonardo Sales

{"title":"关于巴西房地产估价的一点说明","authors":"Thiago Marzagão, R. Ferreira, Leonardo Sales","doi":"10.5935/0034-7140.20210003","DOIUrl":null,"url":null,"abstract":"Abstract Brazilian banks commonly use linear regression to appraise real estate: they regress price on features like area, location, etc, and use the resulting model to estimate the market value of the target property. But Brazilian banks do not test the predictive performance of those models, which for all we know are no better than random guesses. That introduces huge inefficiencies in the real estate market. Here we propose a machine learning approach to the problem. We use real estate data scraped from 15 thousand online listings and use it to fit a boosted trees model. The resulting model has a median absolute error of 8.16%. We provide all data and source code.","PeriodicalId":52490,"journal":{"name":"Revista Brasileira de Economia","volume":"75 1","pages":"29-36"},"PeriodicalIF":0.0000,"publicationDate":"2021-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A note on real estate appraisal in Brazil\",\"authors\":\"Thiago Marzagão, R. Ferreira, Leonardo Sales\",\"doi\":\"10.5935/0034-7140.20210003\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract Brazilian banks commonly use linear regression to appraise real estate: they regress price on features like area, location, etc, and use the resulting model to estimate the market value of the target property. But Brazilian banks do not test the predictive performance of those models, which for all we know are no better than random guesses. That introduces huge inefficiencies in the real estate market. Here we propose a machine learning approach to the problem. We use real estate data scraped from 15 thousand online listings and use it to fit a boosted trees model. The resulting model has a median absolute error of 8.16%. We provide all data and source code.\",\"PeriodicalId\":52490,\"journal\":{\"name\":\"Revista Brasileira de Economia\",\"volume\":\"75 1\",\"pages\":\"29-36\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-07-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Revista Brasileira de Economia\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.5935/0034-7140.20210003\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"Economics, Econometrics and Finance\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Revista Brasileira de Economia","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5935/0034-7140.20210003","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Economics, Econometrics and Finance","Score":null,"Total":0}

引用次数: 0

摘要

摘要巴西银行通常使用线性回归来评估房地产：他们根据面积、位置等特征回归价格，并使用由此产生的模型来估计目标房地产的市场价值。但巴西银行并没有测试这些模型的预测性能，据我们所知，这些模型并不比随机猜测更好。这导致房地产市场效率低下。在这里，我们提出了一种机器学习方法来解决这个问题。我们使用从15000个在线房源中收集的房地产数据，并将其用于拟合增强树模型。所得模型的中值绝对误差为8.16%。我们提供了所有数据和源代码。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A note on real estate appraisal in Brazil

Abstract Brazilian banks commonly use linear regression to appraise real estate: they regress price on features like area, location, etc, and use the resulting model to estimate the market value of the target property. But Brazilian banks do not test the predictive performance of those models, which for all we know are no better than random guesses. That introduces huge inefficiencies in the real estate market. Here we propose a machine learning approach to the problem. We use real estate data scraped from 15 thousand online listings and use it to fit a boosted trees model. The resulting model has a median absolute error of 8.16%. We provide all data and source code.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Revista Brasileira de Economia Economics, Econometrics and Finance-Economics, Econometrics and Finance (all)

CiteScore

0.40

自引率

0.00%

发文量

审稿时长

20 weeks

期刊介绍： A Revista Brasileira de Economia (RBE) é a mais antiga publicação de Economia do Brasil, e a segunda mais antiga da América Latina. Seus fundadores foram Arizio de Viana, o primeiro editor, e Eugênio Gudin, um dos mais influentes economistas da história brasileira. A RBE foi apresentada no seu primeiro número pelo professor Luiz Simões Lopes, em uma Introdução que poderia constar ainda hoje de qualquer número da revista.