利用神经网络预测微观空间经济增长

IF 8.1 1区经济学 Q1 ECONOMICS

American Economic Review-Insights Pub Date : 2022-12-01 DOI:10.1257/aeri.20210422

A. Khachiyan, Anthony Thomas, Huye Zhou, G. Hanson, Alex Cloninger, Tajana Rosing, A. Khandelwal

{"title":"利用神经网络预测微观空间经济增长","authors":"A. Khachiyan, Anthony Thomas, Huye Zhou, G. Hanson, Alex Cloninger, Tajana Rosing, A. Khandelwal","doi":"10.1257/aeri.20210422","DOIUrl":null,"url":null,"abstract":"We apply deep learning to daytime satellite imagery to predict changes in income and population at high spatial resolution in US data. For grid cells with lateral dimensions of 1.2 km and 2.4 km (where the average US county has dimension of 51.9 km), our model predictions achieve R2 values of 0.85 to 0.91 in levels, which far exceed the accuracy of existing models, and 0.32 to 0.46 in decadal changes, which have no counterpart in the literature and are 3–4 times larger than for commonly used nighttime lights. Our network has wide application for analyzing localized shocks. (JEL C45, R11, R23)","PeriodicalId":29954,"journal":{"name":"American Economic Review-Insights","volume":null,"pages":null},"PeriodicalIF":8.1000,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Using Neural Networks to Predict Microspatial Economic Growth\",\"authors\":\"A. Khachiyan, Anthony Thomas, Huye Zhou, G. Hanson, Alex Cloninger, Tajana Rosing, A. Khandelwal\",\"doi\":\"10.1257/aeri.20210422\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We apply deep learning to daytime satellite imagery to predict changes in income and population at high spatial resolution in US data. For grid cells with lateral dimensions of 1.2 km and 2.4 km (where the average US county has dimension of 51.9 km), our model predictions achieve R2 values of 0.85 to 0.91 in levels, which far exceed the accuracy of existing models, and 0.32 to 0.46 in decadal changes, which have no counterpart in the literature and are 3–4 times larger than for commonly used nighttime lights. Our network has wide application for analyzing localized shocks. (JEL C45, R11, R23)\",\"PeriodicalId\":29954,\"journal\":{\"name\":\"American Economic Review-Insights\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":8.1000,\"publicationDate\":\"2022-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"American Economic Review-Insights\",\"FirstCategoryId\":\"96\",\"ListUrlMain\":\"https://doi.org/10.1257/aeri.20210422\",\"RegionNum\":1,\"RegionCategory\":\"经济学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ECONOMICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"American Economic Review-Insights","FirstCategoryId":"96","ListUrlMain":"https://doi.org/10.1257/aeri.20210422","RegionNum":1,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ECONOMICS","Score":null,"Total":0}

引用次数: 3

摘要

我们将深度学习应用于日间卫星图像，以高空间分辨率预测美国数据中收入和人口的变化。对于横向尺寸为1.2 km和2.4 km的网格单元(其中美国县的平均尺寸为51.9 km)，我们的模型在水平上的预测R2值为0.85至0.91，远远超过现有模型的精度，在年代际变化方面的预测R2值为0.32至0.46，这在文献中没有对应值，并且比常用的夜间灯光大3-4倍。该网络在局部冲击分析中具有广泛的应用前景。(凝胶c45, r11, r23)

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

Using Neural Networks to Predict Microspatial Economic Growth

We apply deep learning to daytime satellite imagery to predict changes in income and population at high spatial resolution in US data. For grid cells with lateral dimensions of 1.2 km and 2.4 km (where the average US county has dimension of 51.9 km), our model predictions achieve R2 values of 0.85 to 0.91 in levels, which far exceed the accuracy of existing models, and 0.32 to 0.46 in decadal changes, which have no counterpart in the literature and are 3–4 times larger than for commonly used nighttime lights. Our network has wide application for analyzing localized shocks. (JEL C45, R11, R23)

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

American Economic Review-Insights ECONOMICS-

自引率

0.00%

发文量

期刊介绍： The journal American Economic Review: Insights (AER: Insights) is a publication that caters to a wide audience interested in economics. It shares the same standards of quality and significance as the American Economic Review (AER) but focuses specifically on papers that offer important insights communicated concisely. AER: Insights releases four issues annually, covering a diverse range of topics in economics.