In Search of a Job: Forecasting Employment Growth Using Google Trends

We show that Google search activity on relevant terms is a strong out-of-sample predictor for future employment growth in the US over the period 2004-2018 at both short and long horizons. Using a subset of ten keywords associated with “jobs”, we construct a large panel of 173 variables using Google’s own algorithms to find related search queries. We find that the best Google Trends model achieves an out-of-sample R2 between 26% and 59% at horizons spanning from one month to a year ahead, strongly outperforming benchmarks based on a large set of macroeconomic and financial predictors. This strong predictability extends to US state-level employment growth, using state-level specific Google search activity. Encompassing tests indicate that when the Google Trends panel is exploited using a non-linear model it fully encompasses the macroeconomic forecasts and provides significant information in excess of those.

[1]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[2]  Kasper V. Olesen,et al.  Realizing Correlations Across Asset Classes , 2018, Journal of Financial Markets.

[3]  S. Davis,et al.  Measuring Economic Policy Uncertainty , 2013 .

[4]  J. MacKinnon,et al.  Asymptotic theory and wild bootstrap inference with clustered errors , 2019, Journal of Econometrics.

[5]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[6]  Serena Ng,et al.  Working Paper Series , 2019 .

[7]  J. Stock,et al.  Forecasting Using Principal Components From a Large Number of Predictors , 2002 .

[8]  Matt Taddy,et al.  Text As Data , 2017, Journal of Economic Literature.

[9]  L. Kilian,et al.  How Useful Is Bagging in Forecasting Economic Time Series? A Case Study of U.S. Consumer Price Inflation , 2008 .

[10]  Zhi Da,et al.  In Search of Attention , 2009 .

[11]  D'Amuri Francesco,et al.  Predicting unemployment in short samples with internet job search query data , 2009 .

[12]  Frédéric Karamé,et al.  Can Google Data Help Predict French Youth Unemployment , 2012 .

[13]  Brent Meyer,et al.  Lessons for Forecasting Unemployment in the United States: Use Flow Rates, Mind the Trend , 2015 .

[14]  Peter Molnár,et al.  Google Searches and Stock Returns , 2016 .

[15]  S. Johansen,et al.  The Analysis of Marked and Weighted Empirical Processes of Estimated Residuals , 2019, SSRN Electronic Journal.

[16]  I. Welch,et al.  A Comprehensive Look at the Empirical Performance of Equity Premium Prediction II , 2004, SSRN Electronic Journal.

[17]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[18]  Lei Qi,et al.  Sparse High Dimensional Models in Economics. , 2011, Annual review of economics.

[19]  H. White,et al.  Automatic Block-Length Selection for the Dependent Bootstrap , 2004 .

[20]  Martin M. Andreasen,et al.  Estimating the Price Markup in the New Keynesian Model , 2019 .

[21]  David E. Rapach,et al.  Forecasting US employment growth using forecast combining methods , 2008 .

[22]  Benedikt Maas.  Short‐term forecasting of the US unemployment rate , 2020 .

[23]  Norman R. Swanson,et al.  Forecasting Financial and Macroeconomic Variables Using Data Reduction Methods: New Empirical Evidence , 2010 .

[24]  P. Dupas,et al.  Demand and Welfare Analysis in Discrete Choice Models with Social Interactions , 2019, Review of Economic Studies.

[25]  Francesco Ravazzolo,et al.  Forecasting Macroeconomic Variables Using Disaggregate Survey Data , 2011 .

[26]  J. Oeppen,et al.  Longevity forecasting by socio‐economic groups using compositional data analysis , 2020, Journal of the Royal Statistical Society: Series A (Statistics in Society).

[27]  Zebin Yang,et al.  Online big data-driven oil consumption forecasting with Google trends , 2019, International Journal of Forecasting.

[28]  Erik Christian Montes Schütte,et al.  In Search of a Job: Forecasting Employment Growth Using Google Trends , 2020, Journal of Business & Economic Statistics.

[29]  P. Narayan,et al.  Economic Significance of Commodity Return Forecasts from the Fractionally Cointegrated VAR Model , 2017 .

[30]  Nicolaj Norgaard Muhlbach,et al.  Targeting predictors in random forest regression , 2020, SSRN Electronic Journal.

[31]  Klaus Wohlrabe,et al.  Forecasting with many predictors - Is boosting a viable alternative? , 2011 .

[32]  Matthew D. Webb,et al.  Fast and wild: Bootstrap inference in Stata using boottest , 2018, The Stata Journal: Promoting communications on statistics and Stata.

[33]  Klaus Abberger,et al.  Qualitative business surveys and the assessment of employment -- A case study for Germany , 2007 .

[34]  P. Burridge,et al.  Unit root tests in the presence of uncertainty about the non-stochastic trend , 2000 .

[35]  David Coble,et al.  Now-Casting Building Permits with Google Trends , 2017 .

[36]  Peter Reinhard Hansen,et al.  The Model Confidence Set , 2010 .

[37]  Eric Hillebrand,et al.  The dynamics of factor loadings in the cross-section of returns , 2018 .

[38]  Andrea Gatto,et al.  Defining, measuring and ranking energy vulnerability , 2019 .

[39]  Juri Marcucci,et al.  The Predictive Power of Google Searches in Forecasting Unemployment , 2012 .

[40]  Seth Pruitt,et al.  Market Expectations in the Cross Section of Present Values , 2012 .

[41]  Oscar Claveria,et al.  Business and consumer expectations and macroeconomic forecasts , 2007 .

[42]  H. Varian,et al.  Predicting the Present with Google Trends , 2009 .

[43]  Charlotte Christiansen,et al.  Mutual Fund Selection for Realistically Short Samples , 2019, Journal of Empirical Finance.

[44]  Nadia Vozlyublennaia,et al.  Investor attention, index performance, and return predictability , 2014 .

[45]  Yaojie Zhang,et al.  Forecasting crude oil prices with a large set of predictors: Can LASSO select powerful predictors? , 2019 .

[46]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[47]  Torsten Schmidt,et al.  Forecasting Private Consumption: Survey-Based Indicators vs. Google Trends , 2009 .

[48]  J. Bai,et al.  Forecasting economic time series using targeted predictors , 2008 .

[49]  S. Johansen,et al.  Uniform Consistency of Marked and Weighted Empirical Distributions of Residuals , 2019, SSRN Electronic Journal.

[50]  N. Askitas,et al.  Google Econometrics and Unemployment Forecasting , 2009, SSRN Electronic Journal.

[51]  J. Oeppen,et al.  Forecasting causes of death by using compositional data analysis: the case of cancer deaths , 2018, Journal of the Royal Statistical Society: Series C (Applied Statistics).

[52]  Zhi Da,et al.  The Sum of All FEARS: Investor Sentiment and Asset Prices , 2013 .

[53]  S. B. Thompson,et al.  Predicting Excess Stock Returns Out of Sample: Can Anything Beat the Historical Average? , 2008 .