Department of Economics and Business Economics

In Search of a Job: Forecasting Employment Growth in the US using Google Trends

Research output: Working paperResearch

Standard

In Search of a Job: Forecasting Employment Growth in the US using Google Trends. / Montes Schütte, Erik Christian.

Aarhus : Institut for Økonomi, Aarhus Universitet, 2018.

Research output: Working paperResearch

Harvard

APA

Montes Schütte, E. C. (2018). In Search of a Job: Forecasting Employment Growth in the US using Google Trends. Institut for Økonomi, Aarhus Universitet. CREATES Research Papers, No. 2018-25

CBE

MLA

Montes Schütte, Erik Christian In Search of a Job: Forecasting Employment Growth in the US using Google Trends. Aarhus: Institut for Økonomi, Aarhus Universitet. (CREATES Research Papers; Journal number 2018-25). 2018., 47 p.

Vancouver

Montes Schütte EC. In Search of a Job: Forecasting Employment Growth in the US using Google Trends. Aarhus: Institut for Økonomi, Aarhus Universitet. 2018 Sep 3.

Author

Montes Schütte, Erik Christian. / In Search of a Job: Forecasting Employment Growth in the US using Google Trends. Aarhus : Institut for Økonomi, Aarhus Universitet, 2018. (CREATES Research Papers; No. 2018-25).

Bibtex

@techreport{2a3e3b317a4a4a12b844b767b89044a3,
title = "In Search of a Job: Forecasting Employment Growth in the US using Google Trends",
abstract = "We show that Google search activity on relevant terms is a strong out-of-sample predictor of future employment growth in the US and that it greatly outperforms benchmark predictive models based on macroeconomic, financial, and sentiment variables. Using a subset of ten keywords, we construct a panel with 211 variables using Google{\textquoteright}s own algorithms to find related search queries. We use Elastic Net variable selection in combination with Partial Least Squares to extract the most important information from a large set of search terms. Our forecasting model, which can be constructed in real time and is free from revisions, delivers an out-of-sample R^2 statistic of 65% to 88% for horizons between one month and one year ahead over the period 2008-2017, which compares to between roughly 30% and 60% for the benchmark models.",
keywords = "Forecast comparison, partial least squares, elastic net, complete subset regressions, bagging",
author = "{Montes Sch{\"u}tte}, {Erik Christian}",
year = "2018",
month = sep,
day = "3",
language = "English",
series = "CREATES Research Papers",
publisher = "Institut for {\O}konomi, Aarhus Universitet",
number = "2018-25",
type = "WorkingPaper",
institution = "Institut for {\O}konomi, Aarhus Universitet",

}

RIS

TY - UNPB

T1 - In Search of a Job: Forecasting Employment Growth in the US using Google Trends

AU - Montes Schütte, Erik Christian

PY - 2018/9/3

Y1 - 2018/9/3

N2 - We show that Google search activity on relevant terms is a strong out-of-sample predictor of future employment growth in the US and that it greatly outperforms benchmark predictive models based on macroeconomic, financial, and sentiment variables. Using a subset of ten keywords, we construct a panel with 211 variables using Google’s own algorithms to find related search queries. We use Elastic Net variable selection in combination with Partial Least Squares to extract the most important information from a large set of search terms. Our forecasting model, which can be constructed in real time and is free from revisions, delivers an out-of-sample R^2 statistic of 65% to 88% for horizons between one month and one year ahead over the period 2008-2017, which compares to between roughly 30% and 60% for the benchmark models.

AB - We show that Google search activity on relevant terms is a strong out-of-sample predictor of future employment growth in the US and that it greatly outperforms benchmark predictive models based on macroeconomic, financial, and sentiment variables. Using a subset of ten keywords, we construct a panel with 211 variables using Google’s own algorithms to find related search queries. We use Elastic Net variable selection in combination with Partial Least Squares to extract the most important information from a large set of search terms. Our forecasting model, which can be constructed in real time and is free from revisions, delivers an out-of-sample R^2 statistic of 65% to 88% for horizons between one month and one year ahead over the period 2008-2017, which compares to between roughly 30% and 60% for the benchmark models.

KW - Forecast comparison, partial least squares, elastic net, complete subset regressions, bagging

M3 - Working paper

T3 - CREATES Research Papers

BT - In Search of a Job: Forecasting Employment Growth in the US using Google Trends

PB - Institut for Økonomi, Aarhus Universitet

CY - Aarhus

ER -