Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

OECD Business Cycle Method Leading Indicator

Analysis of Miscellaneous Industry Sector in
Indonesia Stock Exchange

Widodo Prasetyo1 Abitur Asianto2

Student of Magister Management, Lecture of Postgraduate,
Mercubuana University Mercubuana University
Jakarta, Indonesia Jakarta, Indonesia

Abstract:- This research aims to analyze those leading The Miscellaneous Industry Index became one of the
indicator data transformation, leading indicator sectoral that indicated fell during this global financial crisis in
candidate’s selection and its determined of leading 2008. This shows that the ups and downs of macroeconomic
indicator candidates. Monthly price of Miscellaneous conditions abroad will cause shocks in Indonesia's capital. In
Industry Index as the reference series. While the proxy of 2008, when the global financial crisis occurred causing from
leading indicator candidates in form of several indexes for the subprime mortgage scandal in the USA, the miscellaneous
other sectors at IDX, financial sector and other economic Industry Index became one of Industrial sectors which also
indicators that fills the criteria of Organization for experienced shocks and even placed in the third rank of
Economic Cooperation and Development (OECD) started contributors to the negative movement of Sectoral Index in the
during January 2008 - December 2019. OECD business end of December 2008.
cycle method aims to analyze these leading indicator (LI)
that moved ahead from the main index movement. However, in 2013 these Miscellaneous Industry Index
Analysis shows that Nasdaq Composite Index, New York fell again -10% to 1,205.01. Then in 2014 these miscellaneous
Stock Exchange, German Stock Index, French Stock Industry Index increased from 8% to 1,307.07. In 2015 it fell
Market Index, Euro Stoxx 50, Nikkei 225, Shanghai again to 1,057. Meanwhile, in the period from 2016 to 2018
Composite Index, Natural Gas Future Price, BI Rate, US the miscellaneous Industry Index has been increasing
Dollar Exchange Rate, & Money Supply (M2) as the most constantly. Meanwhile, in 2019 these miscellaneous Industry
optimal Composite Leading Indicator with cyclical Index was fell again from -12% to 1,223 compared in 2018.
opposite character from this Miscellaneous Industry The declined which occurred in all sectoral index included the
Index. miscellaneous Industry index in February 2019 was more or
less triggered by the optimism of market players towards
Keywords:- Reference Series, Leading Indicator, negotiations between China and the United States which
Miscellaneous Industry Sector, OECD. declined slightly after the chairman of the US trade
representative issued a statement which too soon to predicted
I. INTRODUCTION the results from US-China neg According to these
phenomenon of miscellaneous industry index fluctuation
The movement of stock prices fluctuated, there were which occurred it is necessary to make a forecasting tool
times when it moved up and sometimes it moved down. One which could be used to predicted these movement of
of the changes in the stock price could be measured by stock miscellaneous Industry Index in the future through analysis of
index. Likewise, these miscellaneous Industry Index, that business cycle indicators. One method that could be used to
included in the Sectoral Index at the IDX that could be used as predicted with analyzing economic indicators. These
measured in the changes and price performance of all stocks identification of economic indicators consists of three
of each industrial sub-sector within. These Miscellaneous categories, which is leading, lagging and coincident indicators.
Industry sector consists of miscellaneous Industry sub-sector, The use of leading indicators to estimate the direction of
textile and garment sub-sector, footwear sub-sector, Cable movement from miscellaneous Industry Index up forward.
sub-sector, and electronics sub-sector. While the lagging indicator is used to see which indicators
move after these miscellaneous Industry Index movements.
Coincident indicator uses to see indicators that move along
with these miscellaneous Industry Index.

IJISRT20DEC124 www.ijisrt.com 167

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Sector Closing Price Changed (%)
Mining Industry 877.68 -2.21%
Property Industry 103.49 -2.03%
Miscellaneous Industries 214.94 -0.41%
Consumer Goods Industry 326.84 1.85%
Manufacturing Industry 236.54 6.04%
Trade Industry 148.33 7.66%
Infrastructure Industry 490.35 11.93%
Agricultural Industry 918.77 14.29%
Financial Industry 176.33 16.85%
Basic Industry 134.99 17.95%
Table 1: Comparison of Sectoral Index’s during Period January - December 2008
Source: Fusion Media Limited (2020) which reprocessed by researchers

These research purpose were to earn a composite index expansion. Besides that, leading indicators will also provide
which moves ahead of the reference series, such as the an early warning system when these miscellaneous industry
miscellaneous Industry Index. What needs to emphasized from Index will experience a turning point, for example from the
this research was the result from these leading indicator contraction stage to the expansion stage. Meanwhile, the sum
produced will only provide an overview in the short term at of growth in these miscellaneous Industry Index on certain
which stage the miscellaneous Industry Index will be placed in period was not the actual purposed of this research nor the use
the future, namely whether it is in a period of contraction or of leading indicators produced.

Figure 1. Comparison of Changes in the Miscellaneous Industry Index in 2019

Source: Fusion Media Limited (2020) which reprocessed by researchers

There were several previous researchers which According to these several existing research, it was
conducted these research that related to business cycle through found that there was not any research that specifically
processing Composite Leading Indicator (CLI) data both in analyzed regarding these leading indicators for miscellaneous
Indonesia and international. Researchers who were carried out industry index at Indonesia Stock Exchange. Aside from that
these research such as Zhang & Zhuang (2002), Kusuma et al. the topic that related to business cycle still become an
(2004), Smirnov (2011), Engemann et al. (2011), Marco interesting to study and leading indicators as part of the
Galegati (2014), Qoyum et al. (2014), Dovolil (2016), Andrea Business Cycle Indicator (BCI) type were believed to have an
et al. (2017), Wahyuningsih & Sumantyo (2017), Nasiri et al. ability as a reliable forecasting tool, so it can be used to
(2017), and Asianto (2018). predict the direction of movement from miscellaneous
industrial index in future notations heading to an agreement
(detik.com, 2019).

IJISRT20DEC124 www.ijisrt.com 168

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
II. LITERATURE REVIEW contraction period. The classical cycles approach estimated
the expansion and contraction period based on their absolute
In 1946, Arthur Burns and Wesley Mitchell in their book values. Meanwhile, the growth cycles approach was
defined that Business cycles are a type of fluctuation found in discovering these turning point based on the calculation of the
aggregate economic activity of nations that organize their long-term trend or in other words the growth cycle was
work mainly in business enterprises. Every business cycle has indicated by these reversal of the direction from cycle and its
two types of turning points, such as peak turning points and long-term trend.
trough turning points. These two turning points will be a
signal if the cyclical movement of indicator would change These business cycle indicators basically a form of
from an expansion period into contraction period or indicator commonly used to predict or forecast future
conversely. These two turning points only could be found by economic conditions. There are three categories of indicators
time series data, in which the deviation from it trends. These that classified based on the type of forecasting produced, such
stages will come and go all the time in a country's economy. as: 1) leading indicators, which are several economic variables
that move ahead from the main economic variables; 2) lagging
Ahmad (1996) were explained that Real Business Cycle indicator, which is a series of economic variables that move
theory could be described as a model which tried to illustrated after the main economic variables changed; and 3) coincident
the aggregate fluctuations in these business cycle through real indicator which is a series of economic variables that move
shocks in the economy, such as productivity and technology. along with variable changed.
Another important assumption used in these Real Business
Cycle theory is regarding to the neutrality of money in the This research framework consists of four approaches.
economy which also applies in the short run, where monetary First, Find out the reference series and indicator proxy series.
policy will not affect real variables such as output and the Second, data normalization needs to process so the data would
unemployment rate. The Leading indicator itself was related to be free from these influences of noise. Then the data
the real business cycle theory which assumed that prices and transformed, interpolated and stabilized so it would free from
wages were flexible even in the short term. the issue of seasonal patterns difference. Third, proxy data for
indicators compared statistically against the reference series of
The use of leading indicators was first driven by the leading indicator data selection. Fourth, leading indicator data
National Bureau of Economic Research (NBER) and has been was examined by classical assumptions and the Best Subset
widely used by countries around the world in predicting Regression to get the optimal leading indicator composition”.
turning points of the business cycle. The idea in using leading
indicators was according to the fact if statistically or the time III. RESEARCH METHODOLOGY
series data consists of these four components of seasonal
factors, cyclical factors, trends and irregular components. This This research used quantitative descriptive method by
method will separate the cyclic component from the other process of data in form of numbers as a tool to analyze and
three components, after which the cyclic component's conduct as research studies, especially regarding what has
behavior was analyzed and compared with the reference been researched by (Kasiram, 2008). The variables used in
series. Business cycle research was initially started by these research consist of the reference series variable and the
Kaminsky, Lizondo and Reinhart in 1998 due to currency candidate variable as composite of predecessor indicator
crises in several countries. At that time, the International (Composite Leading Indicator).
Monetary Fund (IMF) had conducted several early detection
analyses by the Markov Swiching method to predict the The population under this research was all miscellaneous
occurrence of a currency crisis, but the results were deemed Industry Index data in Indonesia Stock Exchange which
inaccurate. The Organization for Economic Cooperation and consists of 49 companies in the period of January 2009-
Development (OECD) then developed a business cycle December 2019. Sampling technique used in this research was
analysis based on leading indicators to predict the business purposive sampling. Sample chosen by OECD method which
cycle turning points. This analysis method was finally applied applied to selected those macroeconomic indicators that will
by various countries to predict the macroeconomics. This be used as candidates for the Composite Leading Indicator
analysis considered as quite comprehensive, flexible, current component based on these following criteria: data availability;
and accurate. economic relevance; and fulfillment of statistical criteria.

The formation of leading indicator index for The leading indicator for these miscellaneous Industry
miscellaneous industry sectors in these research used the Index used the method that developed by Organization for
method that developed by OECD. Basically, the OECD Economic Cooperation and Development (OECD). This
method refers to basic method from the business cycle method refers to the basic method from business cycle which
developed by NBER. To analyze cyclical movements of the developed by The National Bureau of Economic Research
OECD method, a growth cycles approach was used, which is a (NBER). In OECD method, the approach used is growth
modification from real business cycle theory. The growth cycles approach, which in this case has several advantages
cycles approach was recognized to have several advantages compared to classical / traditional cycles. Broadly speaking,
compared to the classical cycles approach. The main the stages that should be done in the OECD method included:
difference between growth cycles and classical cycles determining the reference series and indicator proxies; data
approach lies in the calculation of the expansion period and transformation; detrending and smoothing; determination of

IJISRT20DEC124 www.ijisrt.com 169

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
the reference series turning point; separation of candidate IV. RESULT AND DISCUSSION
leading indicators (LI); leading indicator results were
examined by classical assumptions; and likewise it would be A. Data Transformation
examined by best subset regression (BSR). Determination of turning points in these business cycle
methods is very important, because through this it could be
identified the cause of a recession or shock. The Busy for
Business Cycle 4.1 program with Bry-Boschan method that
used to identify turning points from these miscellaneous
Industry Index for this period of January 2009 - December
2019. Based on these research results, it could be seen that
these miscellaneous Industry Indices consist of three peaks
(peaks) and three peaks troughs.

Figure 2. Transformation Results & Turning Points of Miscellaneous Industry Index Reference Series

Figure 2 above consists of actual data (black chart), turning point analysis; (5) dynamic factor analysis. According
trend (blue graph), and cycle (red graph) Miscellaneous to coherence analysis result, it shows that Indonesian FTSE
Industry Index) during the study period 2009 - 2019. The Index (A12FTSEI), the Manufacturing Index (A8MFG) and
actual data graph is a graph of various industry indices before US Dollar Exchange Rate (H2USD) had the greatest co-
the transformation using Hodrick- Prescote Filter. movement power against these miscellaneous industry Index
Meanwhile, the trend graph is a graph of the trend of the reference series (A1MSIC).
indexes of various industries during the study period, where
at the beginning of 2009 to 2011 there was a significant The results of mean delay analysis shows that there are
upward trend. 12 variables that have an indication of the leading nature of
these miscellaneous industry sector index reference series,
Based on the cycle graph and its turning point, it can be which is the Agricultural Index (A3AGRI), the Mining Index
seen that the index reference series for various industries has (A9MING), the Nikkei 225 Index (D1NKEI), the price of
two long cycles with a cycle duration of 25 months and 33 Brent Oil Future (E1BROIL) Natural Gas Future (E3NGAS)
months respectively, so that the average cycle reaches 29 price, gold future price (F1GLD), silver future price
months as can be seen in Table 4.1. There are seven turning (F2SLVR), nickel future price (F4NCKL), tin future price
points that can be captured by the cyclical movement of (F5TIN), SG X SICOM TSR 20 (G2RUBR), Bank Indonesia
various industry indices, which consist of three troughs and Rate (H1BI ), Export (JEXIM). The mean delay analysis result
four peaks. The average duration of expansion was 14.33 also shows that 21 lagging variables on these miscellaneous
months and contraction was 9.33 months. Industry Sector Index and the remaining 7 variables were
coincident to these miscellaneous Industry Sector Index.
B. Separations of Candidate Components for Composite
Leading Indicator The results from these cross-correlation analysis shows
The separation of CLI component candidates use five that there are 11 variables that are leading to the miscellaneous
econometric methods, such as (1) coherence method; (2) mean Industry Sector Index, such as Nasdaq Index (B2NSDQ),
delay method; (3) cross correlation method; (4) Bry-Boschan Nikkei 225 Index (D1NKEI) and the price of Brent Oil Future

IJISRT20DEC124 www.ijisrt.com 170

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
(E1BROIL). These cross-correlation analysis results also show value indicated that these variable is leading. The positive
that there are 17 variables that are coincident to these variable indicates that the variable moves afterward so it is
miscellaneous Industry Sector Index and the remaining 12 called a lagging variable. Meanwhile, a variable that has zero
variables were lagging on it. value (0) indicated that these variable moves simultaneously
so that it is called a coincident variable. The purpose of this
A turning point analysis was performed after the data turning point analysis detection was to analyze the
series variable was cleaned from seasonal, irregular and trend characteristics of the variables through the reversal behavior of
elements. These turning points analysis were based on the the reference series. Thus, those proxy variables could be
theory which developed by Bry & Boschan (1971). A negative identified whether it is leading, lagging or coincident.

Figure 3. CLI Candidate Variables from Miscellaneous Industry Index

Results from the analysis of common component

variance ratio towards series variance shows that all variables C. Clasic Assumptions Test
have commonality strength of more than 10%. Thus, all The normality test shows that J-B value was 1.577803
variables act as proxy indicators on miscellaneous industry <2 with probability was 0.454344> 0.05. Meaning if the
Index business cycle. Idiosyncratic variables were not found hypothesis H0 which said that these error was normally
so there were no variables removed from these analyses. distributed was accepted. Multicollinearity test by correlation
matrix between independent variables shows that
Cross correlations analysis with dynamic factor method multicollinearity did not represents the resulting model. Its
shows that there has a maximum number in each line that is because the correlation between independent variables was not
bolded and underlined. These numbers were located in that high (above 0.90). The autocorrelation test shows that the
negative lag, zero lag and positive lag. If the maximum Durbin Watson (DW) value was 1,200851, so it could be
number is in positive lag then it's evidence if that variable concluded that these regression models did not have
were leading. If the maximum number is in negative lag, it's autocorrelation. Heteroscedasticity testing by Glejser test
evidence that these variable were lagging. Meanwhile, if the shows that these Prob * R2 value (R-squared) > 0.05, which is
maximum number is in zero lag, it's evidence that the variable 0.5686> 0.05. Thus it could be said if these regression model
was coincident. There are maximum numbers which are did not occur the heteroscedasticity. The R-squared value from
positive and some are negative. A negative number indicated these data processed results amounted to 0.8898742 or 89%.
that the variable has a cyclical opposite behavior with the This shows that these regression equation models would be
cyclic nature of the reference series. If negative number is in a able to explain the variation of these relations between
positive lag then these variables were lagging but cyclical dependent variable by 89%. And the remaining 11% were
opposite to the reference series, for example the BI Rate relations from other independent variables which had impact
(H1BI) variable. If a negative number is in a positive lag then on dependent variables but did not included in these models.
the variable was leading but cyclical opposite towards
reference series, as for example the money supplies variable D. Best Subsets Regression (BSR)
(H3M2). Best Subset Regression analysis is a regression analysis
method used to find out which independent variables will be
The grouping based on its characters has shown that included in the index. From these Best Subset Regression
there are 15 variables which are leading to the reference series. analysis result, there were several alternative Composite
Other than that, the 6 variables were coincident to the Leading Indicators (CLI) will be obtained. The BSR results
reference series and 19 variables were lagging to the reference show if there is the most optimal candidate index with the
series. There are 5 variables which marked by an asterisk (*) largest R-Sq, which is 89.2%. The Mallows Cp value, which is
meaning that they are cyclical opposite to the reference series. 14.3 (less than 17), met these criteria ≤ (p + 1), where p is the
These cyclical opposite consists of 5 leading variables and 1 sum of all variables (16 variables). Meanwhile, the lowest S
lagging variable. Though these variables were stand in the value was 0.34234. According to these BSR analysis, there
opposite direction to the reference series, they had that were 14 candidate variables for the Composite Leading
consistency so they could be used as index constituent Indicator index were also consider as the most optimal against
variables. the miscellaneous Industry Index as a reference series, Such as

IJISRT20DEC124 www.ijisrt.com 171

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Mining (A9MING), Nikkei 225 Index (D1NKEI), Brent Oil The next step was to evaluated these cycle lengths of
Future Price (E1BROIL), RBOB Gasoline Future Price each series. Results from this analysis show that time distance
(E2RBOB ), Natural Gas Future Price (E3NGAS), Gold from peak to peak or trough to trough in the table above was
Future Price (F1GLD), Silver Future Price (F2SLVR), Nickel called cycle spanning. Meanwhile, time distance from peak to
Future Price (F4NCKL), Tin Future Price (F5TIN), Live through or trough to peak to be called phase spanning. If
Cattle Price (G1LCTL), Rupiah Exchange Rate against US these two phases spanning were added, it will become one
Dollar (H2USD *), M2 - Money Supply (H3M2 *), Industrial cycle spanning. These analysis result were shows that there
Production Index (J1PI) and Exports (J2EXIM). were no significant difference between cycle spanning and
the number of phase spanning in each of research variables.
E. Re-run Test
The results from 14 variables that obtained from BSR Moving to next step was analyzed those common
analysis above then were re-processed by Busy Program. The component variance ratio towards variance series. These
reprocessed was carried out on variables which passed the common component variance analysis results shows if all
classical assumption test and BSR. This reprocessing stage variables have ratio above 10% so it could be said that all
includes all business cycle analysis processes consisting of variables met the requirement to strengthen the commonality
log or log transformation, de-trending by Hodrick Prescote and also it was found that there were no idiosyncratic
Filter, smoothing, cross correlation, coherence, mean delay, variables occurred Therefore, There’s no variables were
turning point, till composing composite leading indicator by removed from these analysis.
dynamic factor model.
Further step was to carried out these cross correlation
The coherence analysis results show that all variables analyses by dynamic factor model method contained in these
have an average spectrum of at least 0.08 with maximum of Busy Program. These analysis results shows that the bolded
0.33. Thus it could be said that there were 12 variables that and underlined variables were on negative lag, zero lag and
strong enough co-movement value against these positive lag side. The maximum number located in positive
miscellaneous Industry Index. The results from mean delay log were indicated that the variable was leading, which are
analysis shows that there were seven variables which are A9MING, D1NKEI, E1BROIL, E2RBOB, E3NGAS,
leading to the miscellaneous Industry Index, such as F1GLD, F2SLVR, F5TIN, H2USD, H3M2, J1PI, and
A9MING, E1BROIL, E2RBOB, E3NGAS, F5TIN, G1LCTL, J2EXIM. There are only two variables placed in negative log,
and J1PI. Meanwhile, the remaining 7 variables were which is F4NCKL and G1LCTL that indicated both are
potentially lagging towards these miscellaneous Industry lagging. There were maximum numbers which have positive
Index. The results from cross-correlation analysis shows that and negative values. A negative value was indicated that
there were four variables which have potential to be leading variable has cyclical opposite behavior with cyclical character
towards these miscellaneous Industry Index, such as of the reference series, such as these miscellaneous Industry
A9MING, F2SLVR, H2USD, and J1PI. And the remaining 9 Index. From the positive log side, there are 4 leading
variables have characteristics of coincident and one variable variables which are cyclical opposite to the reference series,
was lagging towards these miscellaneous Industry Index. such as A9MING, D1NKEI, E1BROIL, and F2SLVR.

The Next stage was carried out a turning point analysis Another further step was indexing, which is grouping
of these miscellaneous Industry Index. The negative value in towards its character of each variable against the reference
the results of these turning point analysis as reference series series. These analysis results were indicating that there were
above indicated that these variable moves ahead of the six variables which are leading to the reference series, such as
reference series or known as leading. The positive value in the Nikkei 225 Index (D1NKEI), Brent Oil Future Price
the table shows that the variable moves after these (E1BROIL), RBOB Gasoline Future Price (E2RBOB), Tin
miscellaneous Industry Index reference series so those Future Price (F5TIN), Live Cattle Price (G1LCTL) and
variable called lagging. Meanwhile, the zero value indicated Industrial Production Index (J1PI). A variable that marked
that the variable moves side by side with these miscellaneous with an asterisk (*) was indicated that its cyclical opposite to
Industry Index reference series so then these variable were the reference series. Though these variable were in the
coincident. Average lag turning point results show that there opposite direction, but it still has consistency so still can have
were three potential lagging variables, which are E2RBOB, used as an index building variable. Alongside with that, there
F1GLD and H3M2. Then one variable was coincident, which were four variables which are consider as coincident to the
is J2EXIM. Meanwhile, the remaining 10 potential variables reference series, such as the Natural Gas Future Price
were leading to the miscellaneous Industry Index. This right (E3NGAS) variable, Gold Future Price (F1GLD), Money
indicated that most of these variables were potentially leading Supply (H3M2) and Exports (J2EXIM). Meanwhile, the other
to these miscellaneous Industry Index. Next, the median lag four variables that were lagging, such as Mining Index
turning point results shows if there were three variables (A9MING), Silver Future Price (F2SLVR), Nickel Future
which had potential to be lagging, which are E2RBOB and Price (F4NCKL) and Rupiah Exchange Rate against US
H3M2. The remaining 11 potential variables were leading to Dollar (H2USD).
these miscellaneous Industry Index which shows that the
most of potential variables were leading.

IJISRT20DEC124 www.ijisrt.com 172

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Each leading, coincident and lagging index that has
been processed into composite index by the Busy Program.
The results from these Composite Leading Indicator (CLI)
then compared with the reference series. The results from CLI
graphics towards common component of the miscellaneous
Industry Sector Index (A1MSIC) as reference series was
shown that the red graphics was the Composite Leading
Indicator (CLI) chart proposed by the Busy Program as
composite index of six leading variables. The CLI graphic
consists of the original series index and the results of the
turning point analysis. In these figure, it could be seen that
Table 2. Re-run Variable Nature Index Against the Reference CLI graphics moves ahead of the reference series graph
Series which is the miscellaneous industry Index in which colored
blue.

Figure 4. CLI graphics against miscellaneous industry common components index

F. Disscusion In April 2014, the turning point of the CLI peak coincided
Based on the results of the analysis previously with the turning point of the Miscellaneous Industry Index.
described, the most optimal Composite Leading Indicator Thus, the CLI changed its function to become a coincident in
(CLI) index consists of six variables, namely the Nikkei 225 that period. Therefore, in the next research, it is necessary to
Index (D1NKEI), Brent Oil Future Price (E1BROIL), RBOB carry out further evaluation of the indicators forming the CLI,
Gasoline Future Price (E2RBOB), Tin Future Price (F5TIN), so that their accuracy can be tested again. Then in 2016 the
Live Cattle Price (G1LCTL), and Industrial Production Index CLI moved into a contractionary position four months before
(J1PI). Can be seen in Figure 4.4. above that the CLI the Miscellaneous Industry Index contraction position.
movement is quite good following the cyclical movement of Meanwhile, in 2017, CLI moved into an expansionary
various industry index reference series. position six months before the Miscellaneous Industry Index
expansion position.
In order to be more comparable, the comparison of the
turning points between the various industry index reference The Nikkei 225 Index variable is the main index on the
series and the CLIs formed is presented in Table 4.18. It can Tokyo Stock Exchange (TSE). The Nikkei 225 index is seen
be seen that the CLI is able to capture all turning points as an important barometer for the stock market and the
contained in the various industry index reference series. The economy in Japan. Some even say that the Nikkei 225 Index
lead time averaged 3.4 months for each turning point. The last is the same as the Dow Jones Index in the United States. This
high point of the various industry index reference series index is the oldest stock index in the Asian region because it
captured by the CLI was in June 2018. has existed since 1950. The analysis shows that the Nikkei
225 Index is a leading indicator that is cyclical opposite to the
Based on Table figure 4. It can be seen that in 2011 the reference series, namely the Miscellaneous Industry Index on
CLI moved into an expansion position four months before the the Indonesia Stock Exchange. This shows that the increase
Miscellaneous Industry Index expansion position. In 2012, (decrease) in the capital market index in Japan can be used as
CLI moved into a contractionary position three months before an early warning system for the decline (increase) in the
the contraction position of the Miscellaneous Industry Index. Miscellaneous Industry Index.

IJISRT20DEC124 www.ijisrt.com 173

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
The Brent Oil Future (E1BROIL) variable is a brent oil Tin Future Price (F5TIN) is a metal futures commodity
commodity futures that refers to the Brent Intercontinental traded in the largest and oldest metal trading center in the
Exchange (ICE) index. Brent oil is one of the main energy world, namely the London Metal Exchange. LME has
commodities that are needed by the world. The results of become a reference for the price of metals in the world such
processing brent oil can be used as a source of energy to carry as aluminum, aluminum alloy, cobalt, copper, lead,
out production activities for industries in the world. Brent oil molybdenum, nickel, tin, zinc, and others. Tin is one of the
is a mixture of fifteen types of crude oil originating from the metal commodities most often used in daily life and industrial
North Sea (Lestari et al., 2018). The current brent oil blend purposes. Its soft, malleable, and shiny nature makes it often
consists of crude oil produced from Brent, Forties (added used as a mixture of minerals for various industrial uses such
2002), Oseberg (added 2002), Ekofisk (added 2007), and the as automotive, electricity, food packaging, glass, battery
Troll oilfield (added 2018) also known as BFOET Quotation components, and others. Meanwhile, nickel is often used,
(Imsirovic, 2019). . Brent oil is included in the main trade among others, in the manufacture of stainless steel,
classification of light sweet crude which serves as the automotive frames, wire making, and so on. The results of the
benchmark price for oil purchases worldwide. The level of analysis show that Tin Future commodity prices are a leading
light sweet (light) is assigned to this commodity because of indicator for the Miscellaneous Industry Index on the
its relatively low density and sweetness due to its low sulfur Indonesia Stock Exchange. Thus the increase (decrease) in
content. Brent oil is the leading global price benchmark for the price of Tin Future commodities can be used as an early
Atlantic basin crude. It is even used to determine the price of warning system for the increase (decrease) in the price of the
two thirds of the world's internationally traded supply of Miscellaneous Industry Index.
crude oil. The results of the analysis show that the Brent Oil
Future Price is a leading indicator for the reference series, The Live Cattle Future Price (G1LCTL) variable is a
namely the Miscellaneous Industry Index on the Indonesia live cattle commodity futures traded on the main commodity
Stock Exchange. Thus the increase (decrease) in the price of exchange of the Chicago Mercantile Exchange (CME). Live
the Brent Oil Future can be an early warning system for the Cattle Future Price was first introduced in 1964. This
increase (decrease) in the price of the Miscellaneous Industry commodity future is commonly used as a means of hedging
Index. and for speculating on the price of animal feed. The analysis
shows that the Live Cattle Future Price is a leading indicator
The RBOB variable Gasoline Future Price (E2RBOB) that is cyclical opposite to the Miscellaneous Industry Index
or RBOB gasoline futures contract is a gasoline-fueled on the Indonesia Stock Exchange. Thus, the increase
commodity futures traded on the New York Mercantile (decrease) in the Live Cattle Future Price can be used as an
Exchange (NYMEX). RBOB gasoline is a derivative product early warning system for the increase (decrease) in the price
of refining crude oil which is used as fuel for two-wheeled, of the Miscellaneous Industry Index.
three-wheeled and four-wheeled vehicles. . The results of the
analysis show that the commodity price of RBOB Gasoline The Industrial Production Index (J1PI) variable is an
Future Price is a leading indicator for the Miscellaneous economic indicator that calculates the real production output
Industry Index on the Indonesia Stock Exchange. Thus the from the manufacturing, mining and other manufacturing
increase (decrease) in commodity prices of the RBOB sectors. In Indonesia, the Industrial Production Index is
Gasoline Future Price can also be used as an early warning calculated and published by the Central Statistics Agency
system for the increase (decrease) in the price of the (BPS), while in the United States it is conducted by the
Miscellaneous Industry Index. Federal Reserve Board. The results of the analysis show that
the Industrial Production Index is a leading indicator that is
The results of this study support the opinions of several cyclical opposite to the Miscellaneous Industry Index. The
previous researchers. Faizah et al. (2017) explained that the increase (decrease) in the number of Industrial Production
world oil price has an influence on stock prices. Aseleye et al. Index represents a decrease (increase) in the Miscellaneous
(2019) stated that the variations that occur in oil shocks affect Industry Index. The results of this study support the opinion
most of the macroeconomic variables. Likewise, the of previous researchers. Among them are Rahman & Arfianto
empirical results of Berk & Aydogan (2012) show that (2016), which states that the Industrial Production Index is
changes in oil prices significantly and rationally affected one of the Led Economic Indicators that simultaneously
Turkey's stock market activity during only the third sub- affect stock returns.
period, which started after bad credit 2008. Cunado et al.
(2015) argue that in particular, oil supply shocks have a The use of the OECD method in the early warning
limited impact, while demand shocks driven by global analysis supports several previous studies. Zhang & Zhuang
economic activity have a significant positive effect in the four (2002) used the OECD method to examine the business cycle
Asian countries studied, namely Indonesia, India, Korea and in Malaysia and the Philippines and found that the six
Japan. Alena et al. (2017) also found that the sectoral beta economic indicators from each country were all leading
index that is most affected by oil price shocks as one of the indicators. Pedersen & Elmer (2002) used the OECD method
economic variables is the beta of the agricultural sector, to prove the relationship between the business cycle and
various industries, consumer goods and finance. economic growth. Kusuma et al. (2004) used the OECD
method to analyze short-term leading indicators of investment
in Indonesia with a predictive ability of 1.5 to 4.5 months
ahead. Dovolil (2016) uses the OECD method to predict the

IJISRT20DEC124 www.ijisrt.com 174

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
S&P 500 stock index during the period 2007 - 2014 and finds [5]. Aseleye, A.J., Aremu, C., Lawal, A.I., Ogundipe, A.A.,
that there are three economic indicators in the United States Inegbedion, H., Papoola, O., Sunday, A., & Obasaju,
that are leading indicators. Wahyuningsih & Sumantyo O.B. (2019). Oil Price Shock and Macroeconomic
(2017) using the OECD method with a reference series of Perfomance in Nigeria: Implication on Employment.
GDP and IHSG found that there are several economic International Journal of Energy Economics and Policy.
indicator variables that meet the criteria as leading indicators. 9(5), 451-457. ISSN: 2146-4553.
Asianto (2018) uses the OECD method to analyze the crude [6]. Berk, I., & Aydogan, B. (2012). Crude Oil Price Shock
oil business cycle. WTI found that CLI has an accuracy of up and St'ock Return: Evidence Turkish Stock Market
to 93 percent. Vrana (2018) uses the OECD method to Under Global Liquidity. Institute of Energy Economics
analyze the characteristics of international CLI in Austria, at the University of Cologne (EWI). ISSN: 1862-3808.
Czech Republic, Germany, Poland, and Slovakia where [7]. Bry, G., Boschan, C.. (1971). Cyclical Analysis of Time
economies in Europe are open to each other so that their Series: Selected Procedure and Computer Program.
business cycles are related to the business cycles of other National Bureau of Economic Research. Technical
European countries. Paper, No. 20 (New York).
[8]. Cunado, J., Jo, S., & Garcia, F.Pd. (2015).
V. CONCLUSION Macroeconomics Impact of Oil Price Shock in Asian
Economies. Energy Policy.
This study aims to form a Composite Leading Indicator https://doi.org/10.1016/j.enpol.2015.05.004.
system for various industry index reference series using [9]. Dovolil, J. (2016). The Use of Economic Indicators As a
available macroeconomic indicator data and refers to the Tool for Predicting S&P 500 Stock Index. ACC Journal.
method developed by the OECD, with a view to predicting Vol, 22. Issue 2. DOI: 10.15240/tul/004/2016-2-001.
turning points of various industry index cycles. [10]. Engemann, K.M., Kliesen, K.L., & Owyang, M.T.
(2011). Do oil shocks drive business cycles? some U.S.
After analyzing forty macroeconomic variables available and international evidence. Macroeconomic Dynamics.
from various sources in accordance with the provisions of the 15(3): 498-517. DOI:10.1017/ S1365100511000216.
OECD method, the optimal Composite Leading Indicator [11]. Faizah, N.I., Rachmansah, Y., & Anoraga, P. (2017).
(CLI) for various industries index consists of six leading Analisis Pengaruh Inflasi, Harga Minyak Dunia, dan
indicator variables, namely, Nikkei 225 Index (D1NKEI), Nilai Kurs Dolar (USD/IDR) Terhadap Indeks Harga
Brent Oil Future Price (E1BROIL). RBOB Gasoline Future Saham Gabungan (IHSG) di Bursa Efek Indonesia
Price (E2RBOB), Tin Future Price (F5TIN), Live Cattle Price Periode 2011 – 2014. Journal Magisma, 5(2).
(G1LCTL), and Industrial Production Index (J1PI). [12]. Gallegati, M. (2014). Making Leading Indicators More
Leading: A Wavelet-Based Method for The Construction
Based on the graph of its cyclic movements during the of Composite Leading Indexes. OECD Journal: Journal
2009 - 2019 research period, CLI was able to capture seven of Business Cycle Measurement and Analysis. Vol.
turning points in a various industry index consisting of four 2014/1.
valleys and three peaks. The turning points of expansion and [13]. Kaminsky, G., Lizondo, S., & Reinhart, C. M. (1998).
contraction in the CLI during the study period proved to be Leading indicators of currency crises. Staff Papers 45.1,
moving ahead of the various industry index reference series 1-48.
with an average lead time of 3.4 months, so it can be [14]. Kusuma, W.IGP., Surjaningsih, N., & Siswanto, B.
concluded that CLI is classified as a leading indicator for (2004). Leading Indikator Investasi Indonesia dengan
various industry indices. Thus the CLI formed can be used to Menggunakan Metode OECD. Buletin Ekonomi
predict the future movements of various industry indices in Moneter dan Perbankan. Maret 2004.
expansion or contraction. [15]. Lestari, D.I., Supeno, B., & Junaedi, A.T. (2018).
Pengaruh Suku Bunga, Kurs, Tingkat Inflasi, Harga
REFERENCES Minyak Dunia, dan Harga Emas Dunia Terhadap Harga
Saham Pada Indeks Kompas 100. Di akses pada 15 Juli
[1]. Ahmad, S. (1996). Real Business Cycles: A Survey of 2020. https://osf.io/preprints/inarxiv/egcvq/.
Theories and Evidence. Ekonomi dan Keuangan [16]. Nasiri H., K. Taghizadeh, B. Amiri, & V. Shaghaghi
Indonesia Volume XLIV. Nomor 4. Shahri. (2017). Developing Composite Leading
[2]. Alena, E., Achsani, N.A., & Andati, T. (2017). Dampak Indicators to Forecast Industrial Business Cycles in Iran.
Guncangan Variabel Makroekonomi Terhadap Beta International Journal of Research Industrial Engenering.
Indeks Sektoral di Bursa Efek Indonesia. Jurnal Aplikasi Vol. 6, No. 1. hal 69 – 89.
Bisnis dan Manajemen, 3(3). ISSN: 2528-5149. [17]. Qoyum, A., Miftahussurur, & Matae, A. (2014).
[3]. Andrea, T., Beata, G., & Marcel, B. (2017). The Business Cycle and The Macroeconomics Performance:
Composite Leading Indicator for German Business Evidence of Malaysia and Indonesia. Global Review of
Cycle. Journal of Competitiveness, 9(4), 114 – 133. Islamic Economics and Business. Vol. 2, No. 1. 045 –
DOI: 10.7441/joc.2017.04.08. 068. ISSN 2338 – 7920 (0) / 2338 – 2619 (P).
[4]. Asianto, A .(2018). Analisis Terintegrasi Selling Option
on WTI Futures. Unpublished Dissertation. Bogor:
Institut Pertanian Bogor.

IJISRT20DEC124 www.ijisrt.com 175

Volume 5, Issue 12, December – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[18]. Rahman, M.H., & Arfianto, E.D. (2016). Analisis
Pengaruh Variabel Leading Economic Indicator (LEI)
Dan Coincident Economic Indicator (CEI) Terhadap
Return Saham Jakarta Islamic Index (JII). Diponegoro
Journal of Management. Vol. 5, No. 1. hal 1-14.
[19]. Wahyuningsih, A., & Sumantyo, R. (2017). Analysis of
Indonesia Business Cycle through Composite Leading
Indicator Data Processing for Banking Industry. Jurnal
Keuangan dan Perbankan. 21(4): 589–600. Nationally
Accredited: No.040/P/2014.
[20]. Zhang, W., & Zhuang, J. (2002). Leading Indicator of
Business Cycle in Malaysia and The Philippines. ERD
Working Paper. No. 32. Manila: Asian Development
Bank.

IJISRT20DEC124 www.ijisrt.com 176