Aggregating Google Trends: Multivariate Testing and Analysis
Stephen L. France and
Papers from arXiv.org
Web search data are a valuable source of business and economic information. Previous studies have utilized Google Trends web search data for economic forecasting. We expand this work by providing algorithms to combine and aggregate search volume data, so that the resulting data is both consistent over time and consistent between data series. We give a brand equity example, where Google Trends is used to analyze shopping data for 100 top ranked brands and these data are used to nowcast economic variables. We describe the importance of out of sample prediction and show how principal component analysis (PCA) can be used to improve the signal to noise ratio and prevent overfitting in nowcasting models. We give a finance example, where exploratory data analysis and classification is used to analyze the relationship between Google Trends searches and stock prices.
New Economics Papers: this item is included in nep-big
Date: 2017-12, Revised 2018-03
References: View references in EconPapers View complete reference list from CitEc
Citations Track citations by RSS feed
Downloads: (external link)
http://arxiv.org/pdf/1712.03152 Latest version (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:arx:papers:1712.03152
Access Statistics for this paper
More papers in Papers from arXiv.org
Bibliographic data for series maintained by arXiv administrators ().