Bulletin of Monetary Economics and Banking, Vol. 22, No. 1 (2019), pp. 87 - 102
WHICH VARIABLES PREDICT INDONESIA’S INFLATION?
Susan Sunila Sharma1
1Department of Finance & Centre for Financial Econometrics, Deakin Business School, Australia.
Email: s.sharma@deakin.edu.au
ABSTRACT
We use an exhaustive list of Indonesia’s macroeconomic variables in a comparative analysis to determine which predictor variables are most important in forecasting Indonesia’s inflation rate. We use monthly
Keywords: Macroeconomic variables; Inflation;
JEL Classification: C5; E1.
Article history: |
|
Received |
: December 10, 2018 |
Revised |
: March 1, 2019 |
Accepted |
: March 30, 2019 |
Available online : April 30, 2019
https://doi.org/ 10.21098/bemp.v22i1.1038
88 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
I. INTRODUCTION
There is an enormous body of literature on inflation rate predictability. The literature utilizes a wide range of financial and nonfinancial variables in forecasting inflation rate (see, for instance, Goodhart and Hofmann, 2000; Barr and Campbell, 1997; Forni, Hallin, Lippi, and Reichlin, 2003; Mandalinci, 2017; Salisu and Isah, 2018; among others). The focus of this literature is on alternative models, channels, data sets, and countries to improve the forecasting performance on inflation rate. Thus, empirical evidence on the predictability of inflation rate is mixed. Overall, results are not robust with respect to model specification, sample choice, and countries considered.1
Emerging market volatility poses a problem for central banks in controlling inflation. The target inflation rate is almost never achieved. One step toward achieving inflation close to the target objective is to improve the precision of inflation forecasting. This study aims to examine the factors that successfully predict Indonesia’s inflation rate. In Indonesia, the inflation target is set by the government under the Bank Indonesia Law. However, Bank Indonesia is equally responsible and committed to achieving the inflation target. For instance, in 2001, 2005, and 2008 actual inflation was in excess of the target rate by 6.55%, 11.11%, and 6.06%, respectively. On the other hand, in 2003, 2006, 2009, 2011, 2015, and 2016, the actual inflation rate was less than the target rate, implying that, during these years, inflation targets were achieved. However, the overall statistics do not show that the inflation rate target over the last decade was met consistently.2 Several strands of literature focus on Indonesia’s inflation rate. For instance, some research aims to understand the source of inflation in Indonesia (Siregar and Rajaguru, 2005a, b); other research examines the causal relation between money supply and inflation (Hossain, 2005); and third strand examines the determinants of inflation (Wimanda et al., 2011).
Ramakrishnan and Vamvakidis (2002) examine the inflation process in Indonesia using a multivariate framework, where they regress CPI on 13 variables, including proxies for exchange rate, output gap, wages, and foreign inflation rate. Their data span the period 1980 to 2000. Their conclusion is that exchange rate and the foreign inflation rate are statistically significant predictors of inflation. Sari et al. (2016) propose the backpropagation neural network method to forecast inflation rate. They use monthly
1See Stock and Watson (2001) provides a survey of literature on predictability of inflation rate.
2 Source: Bank Indonesia http://www.bi.go.id/en/moneter/inflasi/data/Default.aspx.
Which Variables Predict Indonesia’s Inflation? |
89 |
|
|
2005 to December 2013. The root mean squared error (RMSE) test statistic is used to examine the accuracy of their forecasting model. They also implement the Sugeno FIS model as a benchmark method. Their results show that the performance of their proposed method is better than the competitor model. Finally, Mandalinci (2017) examines the forecasting performance of inflation rate for nine emerging countries (Chile, India, Indonesia, Malaysia, Mexico, Philippines, South Africa, Thailand, and Turkey). Using quarterly data
The
These issues constitute a research gap on inflation forecasting in Indonesia. Our goal is to fill this gap and construct a
First, our approach follows a bivariate predictive regression framework. We use monthly data for 30 macroeconomic variables. We divide these 30 macroeconomic variables into five groups: (i) three measures of bond yield (separated by maturity, namely, government bond yield at one year (BY1Y), five years (BY5Y), and 10 years (BY10Y)); (ii) four measures of interest rate (separated by maturity: one month (JIBOR1), three months (JIBOR3), six months (JIBOR6), and 12 months (JIBOR12)); (iii) two proxies for monetary aggregates (monetary aggregate (M1 and LM2)); (iv) 12
Second, we use a newly developed estimator proposed by Westerlund and Narayan (WN, 2012 and
90 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
persistency of predictor variables such that, instead of diluting the information contained in predictor variables, we can use the variables in their level form, and
Third, we test for both
Our study contributes to the strand of literature that examines the predictability of Indonesia’s inflation rate. Our approaches produce three main findings. First, we uncover strong evidence of
This paper proceeds as follows. Section II discusses data and methodology. We discuss our main findings in Section. Finally, Section IV sets forth our conclusions.
II.DATA AND METHODOLOGY A. Data Set
We demonstrate the importance of our new predictability model for Indonesia’s inflation rate using 30 macroeconomic variables. These predictor variables are divided into the following five categories:
Which Variables Predict Indonesia’s Inflation? |
91 |
|
|
(i)Three measures of bond yield, separated by maturity: BY1Y, BY5Y, and BY10Y
(ii)Four measures of interest rate, separated by maturity: JIBOR1, JIBOR3, JIBOR6, and JIBOR12
(iii)Two proxies for monetary aggregates: M1 and LM2
(iv)Twelve
(v)Nine financial variables: LBCI, LCAP, LCRI, DY, LDJSI, MCAP, LISI, LCI, and
PER
Our data are taken from Sharma, Tobing, and Azwar (2018). These authors note that the data are extracted from the Global Financial Database and the choice of dataset is based purely on data availability. We provide detailed information on our dataset in Table 1.
Table 1.
Data Description
This table provides detail data description of all variables considered in this study. |
|
|
||
|
|
|
|
|
Variables |
Description |
Date |
No. of obs. |
|
BY1Y |
110 |
|||
BY5Y |
110 |
|||
BY10Y |
110 |
|||
JIBOR1 |
342 |
|||
JIBOR3 |
295 |
|||
JIBOR6 |
330 |
|||
JIBOR12 |
256 |
|||
LM2 |
M2 money supply in natural logarithm |
173 |
||
M1 |
M1 money supply |
124 |
||
LCCI |
Indonesia consumer confidence index in natural logarithm |
201 |
||
LCIC |
Indonesia currency in circulation in natural logarithm |
197 |
||
TD3M |
508 |
|||
LEXP |
Export of goods in natural logarithm |
616 |
||
LER |
Indonesian rupiah per USD in natural logarithm |
617 |
||
IMPPI |
Import price index |
329 |
||
EXPPI |
Export price index |
329 |
||
LIMP |
Imports of good in natural logarithm |
616 |
||
LIP |
Industrial production in natural logarithm |
317 |
||
LR |
Average lending rate for working capital |
366 |
||
PP |
Producer prices (excludes oil) |
544 |
||
FER |
Total foreign exchange reserves (excludes gold) |
570 |
||
LBCI |
Business confidence index in natural logarithm |
190 |
||
LCAP |
Jakarta stock exchange capitalization (value traded, USD) in |
341 |
||
natural logarithm |
||||
|
|
|
||
LCRI |
Indonesia cash return index in natural logarithm |
343 |
||
LCI |
Jakarta stock exchange composite index in natural logarithm |
424 |
||
LDJSI |
Dow Jones Indonesia stock index in natural logarithm |
318 |
||
DY |
Dividend yield |
332 |
||
LISI |
Jakarta stock exchange Islamic index in natural logarithm |
216 |
||
MCAP |
Market capitalization measured as percentage of GDP |
281 |
||
PER |
342 |
|||
INF |
Change in consumer price index |
617 |
92 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
B. Methodology
The starting point for developing our inflation rate forecasting model for Indonesia is to create an extensive dataset of predictors of inflation. To accomplish this, we identify 30 predictor variables. Such an exhaustive list provides a comparative analysis to understand which predictor variables are most important. This is important because it will reveal which variables policy makers should track to forecast inflation. Our predictive regression model will take the following form:
(1)
Here Int is Indonesia’s inflation rate at time t,
We use a newly developed estimator proposed by WN (2012, 2015), the WN- FGLS estimator, to examine the null hypothesis of no predictability.3 The key advantage of the
III. MAIN FINDINGS
This section comprises three subsections. We discuss several key statistical features of predictor variables in first subsection, followed by the main predictability results in second subsection. The final subsection discusses our robustness check.
A. Statistical Features of the Data
To examine the null hypothesis of no predictability, it is essential to first ascertain several commonly known key features of
3We do not discuss the derivation of the
Which Variables Predict Indonesia’s Inflation? |
93 |
|
|
length is attained using the Schwarz information criterion, where we begin with a maximum of 14 lags. We find that the null hypothesis of “unit root” is statistically significantly rejected at the 5% significance level (or better) in 6/30 predictors (JIBOR12, LCCI, LIP, FER, DY and PER), which implies that these six predictor variables follow a stationary process. In other words, our results imply that 80% (24/30) of the considered predictor variables follow a
Table 2.
Unit Root Test Results
This table reports the
a maximum of 14 lags and then use the Schwartz Information Criterion to determine the optimal lag length.
Group |
Variables |
AR (1) |
ADF unit root test |
|||
lag length |
||||||
|
|
|
|
|
||
1 |
BY1Y |
0.8207 |
0 |
0.1318 |
||
1 |
BY5Y |
0.9161 |
0 |
0.4341 |
||
1 |
BY10Y |
0.9172 |
0 |
0.4392 |
||
2 |
JIBOR1 |
0.9539 |
0 |
0.0714 |
||
2 |
JIBOR3 |
0.9638 |
0 |
0.1361 |
||
2 |
JIBOR6 |
0.9754 |
0 |
0.3934 |
||
2 |
JIBOR12 |
0.9789 |
11 |
0.0408 |
||
3 |
LM2 |
1.0009 |
0.6157 |
12 |
0.9995 |
|
3 |
M1 |
0.9952 |
1.6489 |
12 |
0.7670 |
|
4 |
LCCI |
0.9104 |
1 |
0.0063 |
||
4 |
LCIC |
0.9932 |
14 |
0.8651 |
||
4 |
TD3M |
0.9886 |
1 |
0.1571 |
||
4 |
LEXP |
0.9968 |
14 |
0.2955 |
||
4 |
LER |
0.9971 |
9 |
0.3046 |
||
4 |
IMPPI |
0.9891 |
0 |
0.6905 |
||
4 |
EXPPI |
0.9949 |
2 |
0.3533 |
||
4 |
LIMP |
0.9973 |
17 |
0.2022 |
||
4 |
LIP |
1.0028 |
3 |
0.0370 |
||
4 |
LR |
0.9949 |
2 |
0.1097 |
||
4 |
PP |
1.0063 |
1.2217 |
1 |
1.0000 |
|
4 |
FER |
0.9942 |
7 |
0.0010 |
||
5 |
LBCI |
0.9526 |
9 |
0.0694 |
||
5 |
LCAP |
0.9895 |
1 |
0.1273 |
||
5 |
LCRI |
0.9968 |
1 |
0.8819 |
||
5 |
LCI |
0.9985 |
1 |
0.3050 |
||
5 |
LDJSI |
0.9941 |
0 |
0.3828 |
||
5 |
DY |
0.9259 |
0 |
0.0054 |
||
5 |
LISI |
0.9928 |
1 |
0.8247 |
||
5 |
MCAP |
0.9777 |
0 |
0.801 |
||
5 |
PER |
0.8878 |
2 |
0.0000 |
||
|
INF |
0.1538 |
14 |
0.0000 |
94 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
Next, we examine whether our predictor variables are endogenous, in two steps. First, we extract residuals by estimating two models: (i) our predictability model, as represented by Equation 1, and (ii) an AR(1) model of the predictor variable, which takes the following form:
Table 3.
Endogeneity and Heteroskedasticity Test Results
This table reports test results for endogeneity and heteroskedasticity in columns 3 and 4, respectively. The endogeneity test is conducted by regressing the error term from the predictor regression model on the error term from the AR(1) model of the predictor variable. The heteroskedasticity test is performed based on the Lagrange multiplier test, which examines the null hypothesis of “no ARCH” at the lag of 6. We do this by estimating an AR(1) model of all predictor variables. Finally, *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
Group |
Variables |
Endogeneity Test |
Heteroskedasticity Test |
|||
Coefficient |
ARCH (6) |
|||||
|
|
|||||
1 |
BY1Y |
0.133 |
0.1692 |
6.6628 |
0.3532 |
|
1 |
BY5Y |
0.2461** |
0.0230 |
6.5449 |
0.3650 |
|
1 |
BY10Y |
0.2069* |
0.0618 |
4.1442 |
0.6572 |
|
2 |
JIBOR1 |
0.6786*** |
0.0000 |
64.033*** |
0.0000 |
|
2 |
JIBOR3 |
0.7755*** |
0.0000 |
61.404*** |
0.0000 |
|
2 |
JIBOR6 |
0.6729*** |
0.0000 |
62.412*** |
0.0000 |
|
2 |
JIBOR12 |
0.6350*** |
0.0000 |
57.3967*** |
0.0000 |
|
3 |
LM2 |
0.1985 |
22.041*** |
0.0012 |
||
3 |
M1 |
0.5398 |
0.7541 |
8.9027 |
0.1791 |
|
4 |
LCCI |
0.1344 |
0.0299 |
1.0000 |
||
4 |
LCIC |
3.0560*** |
0.0002 |
38.976*** |
0.0000 |
|
4 |
TD3M |
0.6821*** |
0.0004 |
35.610*** |
0.0000 |
|
4 |
LEXP |
4.4346 |
0.2453 |
245,36*** |
0.0000 |
|
4 |
LER |
0.9260 |
85.652*** |
0.0000 |
||
4 |
IMPPI |
0.0009 |
0.0399 |
1.0000 |
||
4 |
EXPPI |
0.1631 |
11.183* |
0.0829 |
||
4 |
LIMP |
0.1624 |
0.9503 |
120.80*** |
0.0000 |
|
4 |
LIP |
0.2650 |
189.34*** |
0.0000 |
||
4 |
LR |
1.3571*** |
0.0000 |
2.9878 |
0.8104 |
|
4 |
PP |
0.2058*** |
0.0000 |
38.135*** |
0.0000 |
|
4 |
FER |
0.5883 |
0.7422 |
206.8*** |
0.0000 |
|
5 |
LBCI |
0.4199 |
42.126*** |
0.0000 |
||
5 |
LCAP |
0.1661 |
58.717*** |
0.0000 |
||
5 |
LCRI |
44.492 |
0.1821 |
287.77*** |
0.0000 |
|
5 |
LCI |
0.0904 |
1.6178 |
0.9513 |
||
5 |
LDJSI |
0.0550 |
35.873*** |
0.0000 |
||
5 |
DY |
0.3377 |
0.1272 |
57.041*** |
0.0000 |
|
5 |
LISI |
0.1184 |
13.978** |
0.0299 |
||
5 |
MCAP |
0.5240 |
0.0177 |
1.0000 |
||
5 |
PER |
0.4762 |
6.9016 |
0.3300 |
Which Variables Predict Indonesia’s Inflation? |
95 |
|
|
Finally, we conduct a heteroskedasticity test. We compute heteroskedasticity by running an AR(1) model of predictor variables and we subject the residuals to an autoregressive conditional heteroskedasticity (ARCH) test at lag of six. The ARCH test is a Lagrange
First, we conclude from our preliminary results that all our predictor variables are highly persistent. Second, our findings suggest that the issue of endogeneity and heteroskedasticity is dependent on the macroeconomic variable used in the predictability model of inflation rate. Thus, overall, our results imply that we cannot ignore these salient features of the data in estimating the predictability model of inflation rate using the 30 available macroeconomic variables for Indonesia. Therefore, this becomes the main motivation for our use of the WN (2012, 2015) predictability model, as it simultaneously accounts for all these three statistical features of
B. Predictability Test Results
Here we discuss results for
Our findings are as follows. First, we note that none of the group 1 (bond yield) predictor variables are found to be statistically significant. Second, all proxies for interbank interest rate (group 2) are found to be statistically significant at the 1% level, which implies JIBOR (irrespective of maturity) is a statistically significant predictor of inflation rate. Third, LM2 significantly predicts inflation rate at the 5% significance level, whereas M1 is reported to be a statistically insignificant predictor from group 3 (monetary aggregate). Fourth, we find that fully 11 (except LCCI) monetary and
96 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
Table 4.
This table reports
Group |
Variables |
Coefficient |
Group |
Variables |
Coefficient |
||
1 |
BY1Y |
0.0192 |
0.8485 |
4 |
EXPPI |
0.0000 |
|
1 |
BY5Y |
0.0503 |
0.6063 |
4 |
LIMP |
0.0000 |
|
1 |
BY10Y |
0.0501 |
0.6105 |
4 |
LIP |
0.0795 |
|
2 |
JIBOR1 |
0.5129*** |
0.0000 |
4 |
LR |
0.2787*** |
0.0000 |
2 |
JIBOR3 |
0.4787*** |
0.0000 |
4 |
PP |
0.0000 |
|
2 |
JIBOR6 |
0.4523*** |
0.0000 |
4 |
FER |
0.0001 |
|
2 |
JIBOR12 |
0.4546*** |
0.0000 |
5 |
LBCI |
0.0807 |
|
3 |
LM2 |
0.0394 |
5 |
LCAP |
0.0000 |
||
3 |
M1 |
0.2143 |
5 |
LCRI |
0.0845* |
0.0980 |
|
4 |
LCCI |
0.2861 |
5 |
LCI |
0.0346 |
||
4 |
LCIC |
0.0685 |
5 |
LDJSI |
0.0002 |
||
4 |
TD3M |
0.1885*** |
0.0000 |
5 |
DY |
0.044 |
0.4347 |
4 |
LEXP |
0.0000 |
5 |
LISI |
0.0077 |
||
4 |
LER |
0.0001 |
5 |
MCAP |
0.2070 |
||
4 |
IMPPI |
0.0000 |
5 |
PER |
0.4006 |
Overall, we conclude from the above discussion that 22/30 macroeconomic variables are statistically significant predictors of Indonesia’s inflation rate. When we consider results as per the five groups, we conclude that none of the group 1 (bond yield) variables are statistically significant, whereas from the other four groups, the majority of the variables significantly predict inflation rate.
Next, we turn to
Which Variables Predict Indonesia’s Inflation? |
97 |
|
|
Table 5.
This table reports results for two measures of
Group |
Variables |
RTU |
OOSR2 |
Group |
Variables |
RTU |
OOSR2 |
1 |
BY1Y |
0.9760 |
4 |
EXPPI |
1.8857 |
||
1 |
BY5Y |
0.9777 |
4 |
LIMP |
1.5891 |
||
1 |
BY10Y |
0.9817 |
4 |
LIP |
1.1211 |
||
2 |
JIBOR1 |
1.4368 |
0.0906 |
4 |
LR |
1.6994 |
|
2 |
JIBOR3 |
1.1155 |
0.3331 |
4 |
PP |
1.5407 |
|
2 |
JIBOR6 |
1.3138 |
0.1146 |
4 |
FER |
1.2298 |
0.0081 |
2 |
JIBOR12 |
1.0090 |
0.3666 |
5 |
LBCI |
1.0011 |
0.0286 |
3 |
LM2 |
1.0620 |
0.1093 |
5 |
LCAP |
1.3150 |
0.0894 |
3 |
M1 |
1.2589 |
5 |
LCRI |
1.6383 |
0.0121 |
|
4 |
LCCI |
1.1700 |
5 |
LCI |
1.0526 |
||
4 |
LCIC |
0.9682 |
5 |
LDJSI |
1.5348 |
0.0266 |
|
4 |
TD3M |
1.0691 |
5 |
DY |
0.9963 |
||
4 |
LEXP |
1.6553 |
5 |
LISI |
0.9925 |
0.0366 |
|
4 |
LER |
1.6011 |
5 |
MCAP |
2.0929 |
||
4 |
IMPPI |
1.4012 |
5 |
PER |
0.9984 |
Contrary to our
From our overall empirical results, we find that
C. Robustness Check
For the sake of completeness, we perform a robustness check of our findings as discussed above. We conduct a robustness test for both
98 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
to ascertain whether our results remain unchanged when h=3 and h=6. We report these results in Table 6. Interestingly, our findings remain unchanged. When h=3 and h=6, we find that 21/30 macroeconomic variables are statistically significant predictors of Indonesia’s inflation rate. The only exception is the predictor variable LBCI, which is statistically insignificant when h=3 and h=6 and is found to be a statistically significant predictor of inflation rate when h=1. Therefore, we conclude that
Table 6.
Robustness Check for
This table reports the WN (2012, 2015)
Group |
Variables |
h=3 |
|
|
h=6 |
|
Coefficient |
Coefficient |
|||||
|
|
|||||
1 |
BY1Y |
0.6632 |
0.7888 |
|||
1 |
BY5Y |
0.8499 |
0.0027 |
0.9787 |
||
1 |
BY10Y |
0.8527 |
0.0142 |
0.8889 |
||
2 |
JIBOR1 |
0.3641*** |
0.0000 |
0.9409*** |
0.0000 |
|
2 |
JIBOR3 |
0.3409*** |
0.0000 |
0.0553*** |
0.0000 |
|
2 |
JIBOR6 |
0.3444*** |
0.0000 |
0.2979*** |
0.0000 |
|
2 |
JIBOR12 |
0.3513*** |
0.0000 |
0.3291*** |
0.0000 |
|
3 |
LM2 |
0.0342 |
0.0748 |
|||
3 |
M1 |
0.1362 |
0.6775 |
|||
4 |
LCCI |
0.0574 |
0.4430 |
0.8881 |
||
4 |
LCIC |
0.0219 |
0.0142 |
|||
4 |
TD3M |
0.1559*** |
0.0004 |
0.0873* |
0.0523 |
|
4 |
LEXP |
0.0000 |
0.0000 |
|||
4 |
LER |
0.0000 |
0.0000 |
|||
4 |
IMPPI |
0.0002 |
0.0065 |
|||
4 |
EXPPI |
0.0001 |
0.0006 |
|||
4 |
LIMP |
0.0000 |
0.0000 |
|||
4 |
LIP |
0.0671 |
0.0302 |
|||
4 |
LR |
0.2561*** |
0.0000 |
0.1923*** |
0.0002 |
|
4 |
PP |
0.0000 |
0.0000 |
|||
4 |
FER |
0.0001 |
0.0000 |
|||
5 |
LBCI |
0.6310 |
0.0313 |
0.6859 |
||
5 |
LCAP |
0.0002 |
0.0071 |
|||
5 |
LCRI |
0.0899* |
0.0831 |
0.0939* |
0.0768 |
|
5 |
LCI |
0.0518 |
0.1035 |
|||
5 |
LDJSI |
0.0001 |
0.0010 |
|||
5 |
DY |
0.0527 |
0.3453 |
0.0012 |
0.9829 |
|
5 |
LISI |
0.0067 |
0.0116 |
|||
5 |
MCAP |
0.3090 |
0.3044 |
|||
5 |
PER |
0.2962 |
0.4435 |
Which Variables Predict Indonesia’s Inflation? |
99 |
|
|
Next, we implement a robustness check for
Table 7.
Robustness Test for
This table reports robustness test results for two measures of
Group |
Variables |
RTU |
OOSR2 |
Group |
Variables |
RTU |
OOSR2 |
1 |
BY1Y |
0.9829 |
0.0102 |
4 |
EXPPI |
1.2326 |
0.1929 |
1 |
BY5Y |
1.0018 |
4 |
LIMP |
1.3631 |
0.2097 |
|
1 |
BY10Y |
1.0116 |
4 |
LIP |
1.0261 |
||
2 |
JIBOR1 |
1.2393 |
0.2527 |
4 |
LR |
1.3855 |
0.1144 |
2 |
JIBOR3 |
1.1057 |
0.3045 |
4 |
PP |
1.1486 |
|
2 |
JIBOR6 |
1.1887 |
0.2557 |
4 |
FER |
1.0058 |
0.1001 |
2 |
JIBOR12 |
1.0253 |
0.3093 |
5 |
LBCI |
1.0088 |
0.0181 |
3 |
LM2 |
0.7591 |
0.4202 |
5 |
LCAP |
1.0898 |
0.2837 |
3 |
M1 |
1.0065 |
5 |
LCRI |
1.1205 |
0.2674 |
|
4 |
LCCI |
1.2208 |
5 |
LCI |
1.0484 |
||
4 |
LCIC |
1.0095 |
0.0743 |
5 |
LDJSI |
1.0239 |
0.2781 |
4 |
TD3M |
1.0243 |
0.0775 |
5 |
DY |
1.0182 |
|
4 |
LEXP |
1.2246 |
0.2752 |
5 |
LISI |
1.0534 |
0.0704 |
4 |
LER |
1.0536 |
0.2686 |
5 |
MCAP |
2.0713 |
|
4 |
IMPPI |
1.5455 |
0.0900 |
5 |
PER |
0.9988 |
0.0030 |
IV. CONCLUDING REMARKS
This paper undertakes an
100 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
often ignored in the literature. We find evidence that 22 macroeconomic variables statistically significantly predict Indonesia’s inflation rate.
Second, we use two measures of
Finally, we consider a robustness test for both
REFERENCES
Barr, D.G., and Campbell, J.Y. (1997). Inflation, Real Interest Rates, and the Bond Market: A Study of U.K. Nominal and
Berg, T., and Henzel, S. (2015). Point and density forecasts for the Euro Area using Bayesian VARs. International Journal of Forecasting, 31,
Caggiano, G., Kapetanios, G., and Labhard, V. (2011). Are More Data Always Better for Factor Analysis? Results for the Euro Area, the Six Largest Euro Area Countries and the UK. Journal of Forecasting, 30,
Clark, T., and Ravazzolo, F. (2015). Macroeconomic Forecasting Performance under Alternative Specifications of
D’Agostino, A., Gambetti, L., and Giannone, D. (2013). Macroeconomic Forecasting and Structural Change. Journal of Applied Econometrics, 28,
Devpura, N., Narayan, P.K., and Sharma, S.S. (2018). Is Stock Return Predictability
Duasa, J., Ahmad, N., Ibrahim, M. H., and Zainal, M. (2010). Forecasting Inflation in Malaysia. Journal of Forecasting, 29,
Forni, M., Hallin, M., Lippi, M., and Reichlin, L. (2003). Do Financial Variables Help Forecasting Inflation and Real Activity in the Euro Area. Journal of Monetary Economics, 50,
Giannone, D., Lenza, M., Momferatou, D., and Onorante, L. (2014).
Goodhart, C., and Hofmann, B. (2000). Asset Prices and the Conduct of Monetary Policy, manuscript, London School of Economics.
Groen, J., Kapetanios, G., and Price, S. (2009). Are
Which Variables Predict Indonesia’s Inflation? |
101 |
|
|
Gupta, R., and Kabundi, A. (2011). A Large Factor Model for Forecasting Macroeconomic Variables in South Africa. International Journal of Forecasting, 27,
Hossain, A. (2005). The Sources and Dynamics of Inflation in Indonesia: An ECM Model Estimation for
Mandalinci, Z. (2017). Forecasting Inflation in Emerging Markets: An Evaluation of Alternative Models. International Journal of Forecasting, 33,
Öğünç, F., Akdoğan, K., Başer, S., Chadwick, M., Ertuğ, D., and Hülagü, T. (2013).
Phan, D.H.B., Sharma, S.S., and Tran, V.T. (2018). Can Economic Policy Uncertainty Predict Stock Returns? Global Evidence. Journal of International Financial Markets, Institutions and Money, 55,
Ramakrishnan, U., and Vamvakidis, A. (2002). Forecasting Inflation in Indonesia. IMF Working paper, No. 02/111.
Salisu, A.A., and Isah, K.O., (2018). Predicting US Inflation: Evidence from a New Approach. Economic Modelling, 71,
Sari, N. R., Mahmudy, W. F., and Wibawa, A. P. (2016). Bankpropagation on Neural Network Method for Inflation Rate Forecasting in Indonesia. International Journal of Advances in Soft Computing and its Applications, 8,
Sharma, S. S. (2016). Can Consumer Price Index Predict Gold Price Returns? Economic Modelling, 55,
Sharma, S.S., Tobing, L., and Azwar, P. (2018). Understanding Indonesia’s macroeconomic data: what do we know and what are the implications? Bulletin of Monetary Economics and Banking, 21,
Siregar, R., and Rajaguru, G. (2005a). Base Money and Exchange Rate: Sources of Inflation in Indonesia during the
Siregar, R., and Rajaguru, G. (2005b). Sources of Variations Between the Inflation Rates of Korea, Thailand, and Indonesia During the Post 1997 Crisis. Journal of Policy Modeling, 27,
Stock, J.H., and Watson, M.W. (2001). Forecasting Output and Inflation: The Role of Asset Prices. Mimeo.
Westerlund, J., and Narayan, P.K. (2012). Does The Choice of Estimator Matter When Forecasting Returns? Journal of Banking and Finance, 36,
Westerlund, J., and Narayan, P.K. (2015). Testing for Predictability in Conditionally Heteroskedastic Stock Returns. Journal of Financial Econometrics, 13,
Wimanda, R. E., Turner., P. M., and Hall, M. J. B. (2011). Expectations and the Inertia of Inflation: The Case of Indonesia. Journal of Policy Modelling, 33, 426- 438.
102 |
Bulletin of Monetary Economics and Banking, Volume 22, Number 1, 2019 |
|
|
This page is intentionally left blank