Revisions analysis of OECD composite leading indicators CLI

1
Revisions analysis of OECD composite leading
indicators (CLI)
Third Joint European Commission OECD Workshop on
Business and Consumer Tendency Surveys
Emmanuelle Guidetti
2
Introduction A leading indicator is a statistic
that signals the movements, up and down, of
future economic activity before they occur, and
it may also give some indication of the magnitude
of those movements. What is the purpose of the
OECD CLI? How is the OECD CLI
constructed? What is the current period
performance of the CLI?
3
What is the purpose of the OECD
CLI?
4
How is the OECD CLI constructed?
5

• What is the current performance of the CLI?
• The aim of the current analysis on CLIs is to
  evaluate the quality of the indicator when they
  are first released in order to
• identify areas where their reliability could be
  improved
• provide further information to users on their
  use for economic analyses.

6

• Why do we have revisions?
• Timeliness/availability
• Frequency
• Smoothness
• Other factors

7

• Size of revisions
• Revision RtLt-Pt

8

• Size of revisions over time

9
Assessment of bias
10
Reliability of 1st estimate as a signal of
short-term expansion or slowdown in economic
activity First measure is the sign of the
movements around 90 of the time, for almost of
the countries, the sign of the initial estimates
of year-on-year growth rates are the same as
those published one month later. This ratio
becomes 95 while talking about zone
aggregates. Second measure is the
acceleration/deceleration test Around 80 of the
time, for around 40 of the countries, the first
estimate of year-on-year growth rates signals no
difference in direction with the 2nd estimates.
This ratio becomes almost 90 while talking about
zone aggregates.
11
Reliability of 1st estimate to provide early
signals of turning points in economic
activity
12
Reliability of 1st estimate to provide early
signals of turning points in economic
activity
13
Conclusion 1st releases of CLI are revised
frequently but the size is rather small for most
countries and negligible for aggregates. No
evidence of bias Improvement in the reliability
of the 2nd estimates 1st and 2nd estimates of
year-on-year growth rates give reliable signals
of approaching cyclical turning points It is not
enough to have timely components they also need
to be smooth to guarantee small revisions. This
further reinforces the argument that smoothness
is probably the most important factor explaining
revisions to the OECD CLI.
14
Investigation on the importance of smoothness of
the component series in the size of revisions of
CLI
15
Investigation on the importance of smoothness of
the component series in the size of revisions of
CLI Business or consumer tendency component
with MCD of 4 or 5 or 6
16
Investigation on the importance of smoothness of
the component series in the size of revisions of
CLI The correlation coefficient between
smoothness and mean absolute revision of first
estimates of the CLIs is rather good and equals
to 0.54. The link is even stronger for second
estimates of the CLIs with an extremely high
correlation coefficient of 0.82. The general
pattern seems to be the higher the MCD value the
higher the mean absolute revision.
17
Investigation on the importance of smoothness of
the component series in the size of revisions of
CLI The revision analysis on Irish CLI
components shows the following
R20.89
18
Investigation on the importance of smoothness of
the component series in the size of revisions of
CLI The revision analysis on German CLI
components is
R20. 94
19
Investigation on the importance of smoothness of
the component series in the size of revisions of
CLI If we plot together Irish and German results
we can get the following graph
R20. 85
20
END