Measuring Uncertainty for New Zealand Using Data‐Rich Approach

• Author: Benjamin Wong,Trung Duc Tran,Tugrul Vehbi
• DOI: http://doi.org/10.1111/1467-8462.12339
• Published date: 01 September 2019
• Date: 01 September 2019

The Australian Economic Review, vol. 52, no. 3, pp. 344–352 DOI: 10.1111/1467-8462.12339

Measuring Uncertainty for New Zealand Using Data‐Rich

Approach

Trung Duc Tran, Tugrul Vehbi and Benjamin Wong*

Abstract

This paper develops two state‐of‐the‐art un-

certainty indices for New Zealand by ex-

ploiting two separate data‐rich environments.

The first index follows the methodology out-

lined in Jurado, Ludvigson and Ng (2015) to

construct an estimate of uncertainty based on

a large New Zealand macro dataset. The

second index is constructed based on freely

accessible and real time Google Trends data

to provide a real‐time and freely‐accessible

measure of uncertainty as in Castelnuovo and

Tran (2017) and Shields and Tran (2018).

Both indices do a reasonable job measuring

uncertainty in New Zealand. VAR evidence

documents significant impacts of uncertainty

shocks on GDP in New Zealand.

1. Introduction

Uncertainty is identified as an important

driver of the business cycles. Heightened

uncertainty is seen to play an important role

in causing recent economic episodes such as

the Great Recession or the European Debt

Crisis. This paper constructs two new state‐of‐

the‐art measures of uncertainty for New

Zealand to study the effects of uncertainty

on the New Zealand economy.

The two measures of uncertainties in this

paper are based on two separate data‐rich

environments. The first index follows the

methodology outlined in Jurado, Ludvigson

and Ng (2015) to construct an estimate of

uncertainty based on a large macro dataset

that monitors the economic condition in

New Zealand. The biggest advantage of this

index is that it is constructed by jointly

considering many time series, therefore

allowing the index to capture only the

unpredictable component of the data.

1

However, this uncertainty index requires a

large set of macroeconomic data that is

only released with lags and subsequent

revisions. That makes such an index incap-

able of monitoring the real‐time uncertainty

status of the economy.

The second index is constructed based on

freely accessible, real time Google Trends

data to provide a measure of uncertainty as in

Castelnuovo and Tran (2017) or Shields and

Tran (2018). The construction is based on the

premise that search terms for uncertainty‐

relatedeventsshouldbehighwhenthelevel

of uncertainty in the economy is high. Being

directly‐observable in real time, the Google

* Tran: University of Melbourne, Victoria 3053 Australia;

Vehbi: Reserve Bank of New Zealand, Wellington 6140

New Zealand; Wong: Monash University, Victoria 3145

Australia. Corresponding author: Tran, email <brian.tra-

n@unimelb.edu.au>. The authors would like to thank

Efrem Castelnuovoand seminar participants atthe Reserve

Bank of New Zealand for helpful feedbacks that greatly

improve this manuscript. The views expressed here are the

views of the authors and do not necessarily reflect the

views of the Reserve Bank of New Zealand.

© 2019 The University of Melbourne, Melbourne Institute: Applied Economic & Social Research,

Faculty of Business and Economics

Published by John Wiley & Sons Australia, Ltd