BIG DATA: USING GOOGLE SEARCHES TO PREDICT THE UNEMPLOYMENT RATE - - PowerPoint PPT Presentation

▶
big data using google searches to predict the
SMART_READER_LITE
LIVE PREVIEW

BIG DATA: USING GOOGLE SEARCHES TO PREDICT THE UNEMPLOYMENT RATE - - PowerPoint PPT Presentation

Feb 15, 2023 139 likes •461 views

BIG DATA: USING GOOGLE SEARCHES TO PREDICT THE UNEMPLOYMENT RATE IN THE EU AIECE MEETING BRUSSELS 6 NOV 2015 JOONAS TUHKURI, ETLA, THE RESEARCH INSTITUTE OF THE FINNISH ECONOMY AND THE UNIVERSITY OF HELSINKI PARTNERS 25 Research

slide-1
SLIDE 1

BIG DATA: USING GOOGLE SEARCHES TO PREDICT THE UNEMPLOYMENT RATE IN THE EU

AIECE MEETING BRUSSELS 6 NOV 2015 JOONAS TUHKURI, ETLA, THE RESEARCH INSTITUTE OF THE FINNISH ECONOMY AND THE UNIVERSITY OF HELSINKI

slide-2
SLIDE 2

PARTNERS

25 ¡Research ¡Ins-tu-ons ¡from ¡Europe: ¡ ¡

slide-3
SLIDE 3

100 B

slide-4
SLIDE 4
  • Unemployment rate (Varian & Choi 2012, Askitas &

Zimmerman 2009, Tuhkuri 2014)

  • Housing market (Brynjolfsson & Wu 2013)
  • Sales (Goel et al 2010, PNAS)
  • Macro indicators (Koop & Onorante 2013)
  • Stock market (Preis et al 2013)
  • Consumption (Vosen & Schmidt 2012)
  • Influenza (Ginsberg et al. 2009, Nature)

LITERATURE

slide-5
SLIDE 5

h6ps://www.etla.fi/en/etlanow-­‑eu28/ ¡ Username ¡and ¡password: ¡etlanow2015 ¡

slide-6
SLIDE 6

ETLAnow Maps

slide-7
SLIDE 7

ETLAnow Forecasts

h6ps://www.etla.fi/en/etlanow-­‑eu28/ ¡ Username ¡and ¡password: ¡etlanow2015 ¡

slide-8
SLIDE 8

ETLAnow Search T erms

slide-9
SLIDE 9

ETLAnow on Twitter

slide-10
SLIDE 10

UNEMPLOYMENT

4 6 8 10 12 Unemployment (%) 2004m1 2006m1 2008m1 2010m1 2012m1 2014m1 2016m1 Time

FINLAND ¡

¡

slide-11
SLIDE 11
  • unemployment benefits
  • unemployment office
  • unemployment claim
  • unemployment compensation
  • unemployment insurance*

GOOGLE INDEX

*In ¡Finnish ¡

slide-12
SLIDE 12

GOOGLE INDEX

FINLAND ¡

¡

slide-13
SLIDE 13
slide-14
SLIDE 14

FRANCE

slide-15
SLIDE 15

PORTUGAL

slide-16
SLIDE 16

BELGIUM

slide-17
SLIDE 17

CROSS CORRELATION

slide-18
SLIDE 18
slide-19
SLIDE 19

GRANGER CAUSALITY

slide-20
SLIDE 20

GRANGER CAUSALITY

slide-21
SLIDE 21

Unemployment Lags Google Index

MODEL

slide-22
SLIDE 22
  • Fit the best model you can using the data you have

(which may often be past values of the time series itself.)

  • Add Google Trends data as an additional predictor
  • See how the out-of-sample forecast improves using

mean absolute error using a rolling window forecast.

  • Particularly interest in turning points since they are the

hardest thing to forecast.

MODEL

*Choi, Hyunyoung, and Hal Varian. "Predicting the present with google trends." Economic

Record 88.1 (2012): 2-9 ¡

slide-23
SLIDE 23

Unemployment rate

MODEL

): log(yt) = β0 + β1log(yt−1) + β2log(yt−12) + et ): log(yt) = β00 + β10log(yt−1) + β20log(yt−12) + β30xt + et

Google Index Lag Seasonal effects

slide-24
SLIDE 24
slide-25
SLIDE 25
slide-26
SLIDE 26
slide-27
SLIDE 27
slide-28
SLIDE 28

PANEL DATA

slide-29
SLIDE 29

PANEL DATA

slide-30
SLIDE 30
  • No improvements using search volumes for

Facebook

  • Results vary between countries
  • Possible solution: better search terms

VARIABLES

slide-31
SLIDE 31
  • Google searches predict unemployment
  • Limited to short-term predictions
  • Value for forecasting purposes episodic
  • Improvements still small
  • But useful for economic forecasting

CONCLUSION

slide-32
SLIDE 32

h6ps://www.etla.fi/en/etlanow-­‑eu28/ ¡ Username ¡and ¡password: ¡etlanow2015 ¡