economic sentiment indicator and confidence indicator in...

Workshop ECFIN, 20. - 21. 11. 2006 1

Economic Sentiment Economic Sentiment Indicator and Indicator and

Confidence Indicator in Confidence Indicator in Services in the new Services in the new

weight System weight System

Dagmar BlahovaEdita Holickova

Statistical Office of the SR

Workshop ECFIN, Brussels, 20. - 21. 11. 2006

2

Workshop November 2005Workshop November 2005

Presentation of experimental calculations of ESI with new weights, included also CI in servicesTime series of indicator in services - only from year 2002We ask Infostat – Institute of Informatics and Statistics for retropolation and extension of time series up to 1997 retrospectivelyIn this work, we try to present results of this reconstruction.


3

The The currentcurrent situationsituationEconomic Sentiment Indicator as composite

indicator is calculated on a monthly base since 1997aggregated from results of processing of BTS in industry, construction, retail trade and consumerscalculated as weighted arithmetic mean of four confidence indicators in industry, construction, retail trade and consumers:IES = a*ICI + b*BCI + c*RCI + d*CCIa, b, c, d = weights


4

Weights Weights of of Confidence Confidence IndicatorsIndicators

ICI - Industry 40 % BCI - Construction 20 % RCI - Retail trade 20 %CCI - Consumer 20 %


5

Experimental calculations Experimental calculations including servicesincluding services

ESI has begun to be calculated at the end of 2004as weighted arithmetic mean from five partial components – confidence indicators in industry, construction, retail trade, servicesand consumer by the following formula:IES = a*ICI + b*BCI + c*RCI + d*SCI + e*CCIa, b, c, d, e = weights


6

New New weightsweights of of Confidence Confidence IndicatorsIndicators

ICI - Industry 40 % BCI - Construction 5 % RCI - Retail trade 5 %SCI – Services 30 %CCI - Consumer 20 %


7

Comparison Comparison of of resultsresultsEconomic sentiment indicator incl. services

100,0

105,0

110,0

115,0

120,0

125,0

130,0

135,0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

2000 = 100

Average 1997 - 2005


8

EconomicEconomic sentimentsentiment indicatorsindicatorsEconom ic sentim ent indicator

80,085,090,095,0

100,0105,0110,0115,0120,0125,0130,0135,0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

ESIN in SR inc.services

(2000=100)

80,085,090,095,0100,0105,0110,0115,0120,0125,0130,0135,0

ESIN in SR w . services

(2000=100)


9

CI in Industry and GDPCI in Industry and GDPConfide nce Indicator in Indus try and GDP (quar te r ly)

-10,0-7,0-4,0-1,02,05,08,0

11,014,017,020,0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

CI in IND

99,0100,0101,0102,0103,0104,0105,0106,0107,0108,0109,0

GDP


10

CI in CI in constructionconstruction and GDPand GDPC o n fid e n ce in d icato r in C o n s tr u ctio n an d GDP

(q u ar te r ly)

-80,0-70,0-60,0-50,0-40,0-30,0-20,0-10,0

0,010,020,0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

CI in BUI

99,0100,0101,0102,0103,0104,0105,0106,0107,0108,0109,0

GDP


11

CI in CI in retail trade retail trade and GDPand GDPC o n fid e n ce in d icato r in Re tail T r ad e an d GDP (q u ar te r ly)

-10,0-5,00,05,0

10,015,020,025,030,035,040,0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

C I in RET

99,0100,0101,0102,0103,0104,0105,0106,0107,0108,0109,0

GDP


12

CI in CI in services services and GDPand GDPConfide nce indicator in Se rvice s and GDP (quar te r ly)

20,025,030,035,040,045,050,055,060,065,070,0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

CI in SER

99,0100,0101,0102,0103,0104,0105,0106,0107,0108,0109,0

GDP


13

Consumer CI and GDPConsumer CI and GDP

C o n s u m e r ´s co n fid e n ce in d icato r an d GDP (q u ar te r ly)

-50,0-45,0-40,0-35,0-30,0-25,0-20,0-15,0-10,0

-5,00,0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

CCI

99,0100,0101,0102,0103,0104,0105,0106,0107,0108,0109,0

GDP

Workshop ECFIN, 20. - 21. 11. 2006 14

ConclusionsConclusionsFrom January 2007

new publication of official results of IES with inclusion of confidence indicator in servicesbased on reconstructed time series from 1997 –2001 and on real data from 2002 – 2006contemporaneously with previous weights we will maintain the original time series of IES some period of time. we release new weight system and methodology of calculationwe will inform economic public about this measures in advance

Workshop ECFIN, Brussels, 20. -21. 11. 2006

15

Composite indicators Composite indicators and econometric and econometric

modelsmodels

Jan HaluskaInfostat


16

Research and model Research and model toolstools --INFOSTATINFOSTAT

In Slovak statisticsthe research in regard to the construction and use of models has for a long time been carried out by INFOSTAT. From the methodological point of view there are econometric models.

Since 2005, their development has received a further stimulus in the fact that the SO SRis obliged to compute and publish in advance a flash estimate of GDP and total employment – always by 45 days after the end of each quarter. It is in this context that model tools for computing flash estimates of these indicators were created and applied during the course of 2005.ESI is considered as a reference indicator for its growth (compared to the same period of the previous year).


17

Modelling Modelling of GDP of GDP based based on ESIon ESI

As the starting hypothesis about the functional form of the model relationship can be formulated differently, model tools presented here have different form as well.

Either it is a classical econometric model (not affected by error correction term) or it is a model relationship in the form of ECM.


18

ModellingModelling of GDPof GDP basedbased on ESIon ESI

The results of estimation are based on original (not seasonally adjusted) time series of relevant quantitative and qualitative indicators in combination with seasonal dummies the quarterly time series of the ESI was created by transformation from its original, i.e. monthly time series parameters of model relationships are estimated by means of methods using EViews 5


19

Correlation between Correlation between ESI and ESI and GDPGDP

- 2 8

- 2 4

- 2 0

- 1 6

- 1 2

- 8

- 4

0

4

8

- 1

0

1

2

3

4

5

6

7

8

9 6 9 7 9 8 9 9 0 0 0 1 0 2 0 3 0 4 0 5

G D P g r o w t h , % ( R H S s c a le )E c o n o m ic s e n t im e n t i n d ic a t o r , b a la n c e ( L H S s c a le )


20

ModelModel relationship without error relationship without error correctioncorrection termterm

Based • on the starting hypothesis that the ESI is considered as a

reference indicator for GDP in constant prices • to be more precise, it is assumed that there exists statistically

significant dependency between the percentage growth rate of GDP (compared to the same quarter of the previous year) and the ESI

• the hypothesis formulated in this way was tested using quarterly time series of these indicators for the period 1st quarter 1996 to 4th quarter 2005, i.e. from 40 observations. Figure 1 presented below shows that a close relationship exists between the evolution of the ESI (expressed in balances) and the growth of GDP (in %):


21

Model Model relationship without error relationship without error correction correction termterm

- 2 8

- 2 4

- 2 0

- 1 6

- 1 2

- 8

- 4

0

4

8

- 1

0

1

2

3

4

5

6

7

8

9 6 9 7 9 8 9 9 0 0 0 1 0 2 0 3 0 4 0 5

G D P g r o w t h , % ( R H S s c a le )E c o n o m ic s e n t im e n t in d ic a t o r , b a la n c e ( L H S s c a le )


22

Model Model relationship relationship in ECM in ECM formform

The construction of the model relationship is based on a modified starting hypothesis according to which GDP grows at a basically constant pace, however, under the effect of changes in the ESI, its pace becomes variablethe hypothesis so formulated results in a following model relationship that is assumed to be long-term one:GDP = α * e b * TIME + c * ESI

orlog (GDP) = a + b * TIME + c * ESI


23

ModelModel relationshiprelationship in ECMin ECMformform

- . 0 6

- . 0 4

- . 0 2

. 0 0

. 0 2

. 0 4

. 0 6

4 . 9

5 . 0

5 . 1

5 . 2

5 . 3

5 . 4

5 . 5

9 6 9 7 9 8 9 9 0 0 0 1 0 2 0 3 0 4 0 5

R e s id u a l A c t u a l F i t t e d


24

ConclusionsConclusionsThe ESI

can be considered as statistically significant indicator of GDP development it may be used to construct model relationships for flash estimates of GDP a comparison of the statistical characteristics shows their different explanatory power, significantly higher for the ECM modelto make GDP flash estimates more reliable, depends above all on expanding the ESI with the confidence indicator in services the services sector accounts for more than 50% of GDP in Slovakia


25

ESI, CI and ESI, CI and modelsmodels in Slovakiain Slovakia

Thank you for your attention

economic sentiment indicator and confidence indicator in...

Documents