AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD...
Transcript of AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD...
ANALYSIS OF FACTORSAFFECTING GOLD PRICES
-Dr. Rahul Bishnoi-Jia Lan
ABSTRACTThis paper analyses the critical factors affecting the price of gold
using ordinary least square, white-test and weighted least squares. Yearly datafrom 1994 to 2013 were carefully collected and statistically tested. The resultsshowthat Gold prices, US dollar to Indian Rupee exchange rate, and Crudeoil prices are positivelycorrelated albeit a negative relationshipclearly emergeswith the Rate of Inflation, long run interest rates in the US and their RealGDP.
KEYWORDS:GDP, INFLATION
1. INTRODUCTION2014 saw a decline in gold prices, plus the Modi government is widely
expected to allow more imports of gold over time. This directly affects the Indianconsumers who happen to be among the top buyers of this precious metal. Ourpaper addresses the issue of gold pricing in relation to its most important variablesto help the readers make educated decisions about their future purchases.
Gold, one of the most used investment and hedging tools in the market, hasseen a price increase from 1994 to 2013 on the whole as well as fluctuations in theshort run. Its price increased from an average of US$385.42 per ounce in 1994 toan average of $1700 per ouncein2012.Even taking the inflation into consideration,the gold price still jumped dramatically but it fell to $1598 per ounce from 2012 to2013 and is continuing its downward spiral. (please see the figure below)
Gold price is determined by many factors, such as the US dollar to IndianRupee exchange rate, Inflation Rate, Crude oil prices, US dollar short term andlong terminterest rates, and US real GDP. This paper empirically analyses thesefactors affecting the price of gold by using Ordinary Least Square, Weighted LeastSquare and White-test etc. to help explain this correlation and give guidance toinvestors regarding any future investment that they may have in mind.
AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD PRICES 1
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
AN EMPIRICAL
Interested readers can benefit from the following background.
2. LITERATURE REVIEWMany scholars have shown their interest in gold pricing. A comprehensive
list of 25 scholarly papers can be found in the references to feed the interest of anavid researcher. Capie et al., F. Capie, T.C. Mills, G. Wood (2005) show that al-though gold has served as a hedge against ûuctuations in the foreign exchange rateof the dollar, it has only done so to a degree that seems highly dependent on unpre-dictable political attitudes and events.
Muhammad Shahbaza, Mohammad Iqbal Tahirb, Imran Alia, Ijaz UrRehmand (2014) have applied the ARDL bounds testing approach to co-integrationfor the long run, and innovative accounting approach (IAA) to examine the direc-tion of causality in variables to reveal that “investment in gold is a good hedgeagainst inflation” not only in the long-run but also in the short-run.
Roger C. Van Tassel(1981) found that the use of gold as a major commod-ity explains the long-term real price increase and restrains purely panic-speculativedeviations from a trend.
Shahriar Shafieea and Erkan Topalb (2010)estimate gold prices for thenext 10 years, based on monthly historical data of nominal gold price quotations.
Mark Joy( 2011) has three key findings: (i) During the past 23 years goldhas behaved as a hedge against the US dollar. (ii) Gold has been a poor safe havenand (iii) In recent years, gold has increasingly acted as an effective hedge againstcurrency risk associated with the US dollar.
Joscha Beckmanna and Robert Czudaj (2013) ponder over the question ofgold providing the ability of hedging against inflation from a new perspective, usingdata for four major economies, namely the USA, the UK, the Euro Area, andJapan.
2 IJHPD VOL. 3 NO. 2 JULY - DEC 2014
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
Christian Pierdzioch, Marian Risse and Sebastian Rohloff (2014) find thatthe gold market is informationally efficient with respect to international business-cycle fluctuations.
Li Lili and Diao Chengmei (2013) show the dynamics of gold pricing in theNew York Gold Exchang using a dataset that includes global macroeconomic indi-cators, financial market indices, quantities and prices of energy products. They finda negative correlation with financial market indices and macroeconomic indicatorswhereas the effect of gold reserve and prices of energy product to gold is positive.
Chia-Lin Changa, Jui-Chuan Della Changb and Yi-Wei Huangc (2013)ex-amine the inter-relationships among gold prices in five global gold markets, namelyLondon, New York, Japan, Hong Kong (which became a Special AdministrativeRegion of China on July 1, 2007), and Taiwan.
Yu Long (2013) finds the logarithmic gold price time series to appear as amultifractal Brownian series, the return series of log-gold price appears as amultifractal Gaussian noise series, the visibility graphs of price series and returnseries are both small world networks and the price series is a hierarchy structure inagreement with Elliot’s Wave Theory.
Hsiao-Fen Changa, Liang-Chou Huangb and Ming-Chin Chinc (2013) ex-amine the correlations of oil prices, gold prices and the NT dollar versus U.S. dollarexchange rate during 2007/09/03–2011/12/28.
3. DATA AND METHODOLOGYThis paper usesyearly data from 1994 to 2013 coming mainly from the
MeasuringWorth and OANDA database.Dependent Variable Gold price
US dollar to Indian Rupee exchange rateInflation RateCrude oil price
Independent Variables US dollar short term interest rateUS dollar long term interest rateUS real GDP SP 500 Index
The data for all the variables are presented in table 1. Table 2 shows thedescriptive statistics for all the variables.
(1) Multivariate regression model - If we define gold price as Y, USdollar to Indian Rupee exchange rate as X1, Inflation Rate as X2, crude oil price asX3, US dollar short and long term interest rate as X4and X5, US real GDP as X6, SP500 Index as X7 then the regression model is:
Y= b0 + b1X1 + b2 X2 + b3 X3 + b4 X4 + b5 X5 + b6 X6 + b7 X7
AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD PRICES 3
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
Using Eviews Ordinary Least Square Method, the regression equation isY=4354.772+14.46697X1 -11054.93X2 +21.50028X3-0.121358X4 -
161.0319X- 0.00029798X6+-106.2007X7. (Regression results are shown in Table 3)(2) Statistical Test - R2 is a statistical measure of how well the regression
line approximates the real data points. A larger R2shows a better fit of the regres-sion model to the sample observations.
Adjusted R2 is a modification of R2 that adjusts for the number of explana-tory terms in a model. Unlike R2, the adjusted R2increases only if the new termimproves the model more than would be expected by chance. The adjusted R2 canbe negative, and will always be less than or equal to R2. The larger the adjusted R2,the better the fit of the regression model to the sample observations.
In our regression model, R-squared = 0.966970, adjusted R-squared=0.947703.which exemplifies a good fit to the sample observations. We can also seefrom Table 4, depicting the actual sample, regression model and residual, that theregression model and the sample observations are quite a match.
(3) F-test - The null hypothesis is: H0 :
The alternative hypothesis is that at least one bi does not equal zero(i=1,2,3,4,5,6,7)
We get F-statistic=50.18677 for a significance level of a=0.05. The criticalvalue for a 7-numerator freedom and 22-denominator freedom is 2.464 which isless than 50.18677 so we reject H0 showing that the regression model is indeedsignificant.
(4) White-test - The auxiliary regression result isY=77571.36-24 .23448X 1+5399095+8.033133X 3-1175.637X4
+477.7411X50.000000000178 X6 -4661.099X7with anR2=0.825367(Regression results are shown in Table 5)The white nR2=20*0.825367= 16.20734. From c2 distribution table, we get
that for a significance level of a=0.05, and freedom level=7, thisc2 critical valueisc2
a=0.05=14.067. Because nR2>c2a=0.05it rejects the null hypothesis of
homoscedasticity in this model. Thus to eliminateheteroscedasticity, we use theWeighted Least Squares method)The WLS equation output is Y=3848.471-
13.70254X1 - 9301.900X2+ 21.87139X3 -16.53885X4-111.7561X5 - 0.000278X6 -121.6887X7R2=0.825367(Result shown in Table 6)(6) White test for WLSThe aux-iliary regression result is shown in table 7.HereR2=0.388622, the whitenR2=20*0.388622= 7.77324. From c2 distribution table, we get that for a signifi-cance level of a=0.05, and freedom level=7, the c2 critical value isc2
a=0.05= 14.067.Because nR2<c2
a=0.05 we accept the null hypothesis of Homoscedasticity. So
4 IJHPD VOL. 3 NO. 2 JULY - DEC 2014
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
theheteroscedasticity is eliminated.
4. RESULTS AND CONCLUSIONSFrom the above regression equation and correlation coefficient tests, we
can arrive at the conclusion that at a significance level of 0.05, US dollar to IndianRupee exchange rate, US Inflation Rate, Crude oil price, US dollar long term inter-est rate and US real GDP do have an influence on gold price since the p-value isless than 0.05 for these factors. Furthermore, gold prices toUS dollar to IndianRupee exchange rate, and Crude oil prices have positive relationship whereas thereis negative relationship with Inflation, US long term interest rate, and US real GDP.
5. SOME REFLECTIONS ARE IN ORDER HERE1. As the Rupee exchange rate increases,India will have more demand for
gold in order to keep the value of their money that will result inan increasein gold price.
2. From a historical perspective, for each unit increase in US dollar to IndianRupee exchange rate, the price of gold decreases by 9301.900units. Thehigher the inflation rate, the higher the pricewill beand purchasing power ofmoney will decline, resultingin the gold price to rise significantly.
3. Gold price and crude oil price have positive relationshipas the world’s ma-jor oil spot and futures prices are priced in dollars. Fluctuations in oil pricesrelate to the dollar, thus oil prices and gold prices indirectly influence eachother.
4. Gold prices and US dollar long term interest rate have negative relationssince for each unit US dollar to Indian Rupee exchange rate increase, theprice of gold decreases by 16.53885units. A possible explanation is thatwhen the interest rate is low, the investment in gold will have certain ben-efits, and as the demand for gold increases, the price of gold will increaseconsequentially.
5. Gold prices and US real GDP have negative relationship since historically,for each unit of US dollar to Indian Rupee exchange rate increase, theprice of gold decreases by -0.000278units. The higher the GDP, the betterthe economic conditions, and it will enhance people’s desire to buy gold. Asa result, the high demand increases gold price.From the correlation coeffi-cient matrix (Table 8), we can see the existence ofmulticollinearity for USreal GDPas a factor (correlation coefficients between US real GDP andGold prices, and US dollar to Indian Rupee exchange rate are 0.7299 and0.80567 respectively). Therefore, the regression coefficients of the inter-est rate variable and the dollar index may be somewhat compromised.Further studies can try to overcommulticollinearityby using inflation ad-
justed data or by using monthly or daily data instead.
AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD PRICES 5
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
REFERENCES: AntoninoParisi, Franco Parisi, David Diaz. Forecasting gold price changes: Rolling and
recursive neural network models, Journal of Multinational Financial Management 2008,18:pp. 477-487.
Baur and Lucey, D.G. Baur, B.M. Lucey : Is gold a hedge or a safe haven? An analysis ofstocks, bonds and gold, Financial Review, 2010, pp. 217–229.
Beckmann and Czudaj, J. Beckmann, R. Czudaj, Gold as an inflation hedge in a time-varyingcoefficient framework, North American Journal of Economics and Finance, 24 (2) 2013, pp.208–222
Blose, L.E. Blose, Gold prices, cost of carry, and expected inflationJournal of Economicsand Business, 62 (1) 2010, pp. 35–47
Capie et al., F. Capie, T.C. Mills, G. Wood, Gold as a hedge against the dollar, Journal ofInternational Financial Markets, Institutions and Money, 15 (4) 2005, pp. 343–352
Chia-Lin Changa, Jui-Chuan DellaChangb and Yi-Wei Huangc, Dynamic price integration inthe global gold market. The North American Journal of Economics and Finance, Volume 26,December 2013, Pages 227–235.
Christian Pierdzioch, Marian Risse and Sebastian RohloffThe international business cycleand gold-price fluctuations, The Quarterly Review of Economics and Finance, 2010
Chua and Woodward, J. Chua, R.S. Woodward (1982), Gold as an inflation hedge: a com-parative study of six major industrial countries, Journal of Business Finance and Account-ing, 9 (2) (1982), pp. 191–197
Hsiao-Fen Changa, Liang-Chou Huangb and Ming-Chin Chinc, Interactive relationshipsbetween crude oil prices, gold prices, and the NT–US dollar exchange rate—A Taiwanstudy, Energy Policy,Volume 63, December 2013, Pages 441–448.
JoschaBeckmanna and Robert Czudaj, Gold as an inflation hedge in a time-varying coeffi-cient framework, The North American Journal of Economics and Finance Volume 24, Janu-ary 2013, Pages 208–222.
Kolluri, B.R. Kolluri, Gold as a hedge against inflation: An empirical investigation, Quar-terly Review of Economics and Business, 1981, pp. 13–24.
Koutsoyiannis, A. Koutsoyiannis, A short-run pricing model for a speculative asset, testedwith data from the gold bullion market, Applied Economics, 15 (5) 1983, pp. 563–581
Li Lili and DiaoChengmei, Research of the Influence of Macro-Economic Factors on thePrice of Gold, Procedia Computer Science, Volume 17, 2013, Pages 737–743.
Mark Joy, Gold and the US dollar: Hedge or haven? Finance Research LettersVolume 8,Issue 3, September 2011, Pages 120–131.
Mishra and Rahman, B. Mishra, M. Rahman , The dynamics of Bombay stock, US stock andLondon gold markets, Indian Journal of Economics and Business, 4 (1) 2005, pp. 151–160
Muhammad Shahbaza, Mohammad Iqbal Tahirb, c, Imran Alia, Corresponding author con-
6 IJHPD VOL. 3 NO. 2 JULY - DEC 2014
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
tact information, Ijaz Ur Rehmand, Is gold investment a hedge against inflation in Pakistan?A co-integration and causality analysis in the presence of structural breaks, The NorthAmerican Journal of Economics and Finance, 2014
Moore, G.H. Mooreb, Gold prices and a leading index of inflation Challenge, 33 (4) 1990,pp. 52–56
Qin sheng and Chenyang. Gold price prediction based on Grey Markov chain model,Economists, 2010,10: 29-32.
Roger C. Van Tassel(1981), The changing commodity and investment prospects for gold.Precious Metals 1981, Proceedings of the Fifth International Precious Metals InstituteConference, Held in Providence, Rhode Island, June ‘81
ShahriarShafieea and ErkanTopalb (2010), An overview of global gold market and gold priceforecasting, Resources Policy, Volume 35, Issue 3, September 2010, Pages 178–18.
Sjaastad and Scacciavillani, L.A. Sjaastad, F. Scacciavillani (1996), The price of gold and theexchange rate, J. Int. Money Finance, 15 (1996), pp. 879–897
Tully and Lucey, E. Tully, B.M. Lucey (2007), A power GARCH examination of the goldmarket, Research in International Business and Finance, 21 (2007), pp. 316–325
Wang et al., K.M. Wang, Y.M. Lee, T.B.N. Thi (2011),Time and place where gold acts as aninflation hedge, An application of long-run and short-run threshold model, Economic Mod-elling, 28 (2011), pp. 806–819
Worthington and Pahlavani, A.C. Worthington, M. Pahlavani (2007), Gold investment as aninflationary hedge: Cointegration evidence with allowance for endogenous structural breaks,Applied Financial Economics Letters, 3 (2007), pp. 259–262
Yu Long (2013 ) Visibility graph network analysis of gold price time series, Physica A:Statistical Mechanics and its Applications, Volume 392, Issue 16, 15 August 2013, Pages3374–3384.
Dr. Rahul Bishnoi and Jia LanFrank G. Zarb School of Business
Hofstra University,New York, USA
AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD PRICES 7
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
APPENDIXTABLE 1
DATA FOR ALL VARIABLES
TABLE 2DESCRIPTIVE STATISTICS
8 IJHPD VOL. 3 NO. 2 JULY - DEC 2014
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
TABLE 3ORDINARY LEAST SQUARE TEST
Dependent Variable: GOLD_PRICEMethod: Least SquaresDate: 04/9/14 Time: 14:13Sample: 1994 2013Included observations: 20
Variable
CUS_EXCHANGE_RATE
INFLATION_RATECRUDE_OIL_PRICE
SHORT_TERM_INTEREST_RATELONG_TERM_INTEREST_RATE
REAL_GDPSP500
R-squaredAdjusted R-squaredS.E. of regressionSum squared residLog likelihoodF-statisticProb(F-statistic)
Coefficient
4354.77214.46697-11054.9321.50028-0.121358-161.0319-0.000298-106.2007
0.9669700.947703111.0767148056.3-117.474950.186770.000000
Std. Error
1304.0499.2461104095.6872.73735822.7547871.813546.30E-05101.6723
Mean dependent var S.D. dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter. Durbin-Watson stat
t-Statistic
3.3394241.564655-2.6991647.854392-0.005333-2.242362-4.733389-1.044539
Prob.
0.00590.04360.01930.00000.99580.04460.00050.3168
670.1760485.716812.5474912.9457912.625241.767398
AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD PRICES 9
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
TABLE 4THE ACTUAL SAMPLE, REGRESSION MODEL AND
RESIDUAL
10 IJHPD VOL. 3 NO. 2 JULY - DEC 2014
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
Variable
C
US_EXCHANGE_RATE^2
INFLATION_RATE^2
CRUDE_OIL_PRICE^2
SHORT_TERM_INTEREST_RATE 2̂
LONG_TERM_INTEREST_RATE 2̂
REAL_GDP^2
SP500^2
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)
Coefficient
77571.36
-24.23448
5399095.
8.033133
-1175.637
477.7411
-1.78E-10
-4661.099
0.825367
0.723498
12786.62
1.96E+09
-212.3936
8.102246
0.000944
Std. Error t-Statistic
64335.95 1.205723
10.97825 -2.207499
11666418 0.462789
1.793736 4.478437
409.6542 -2.869828
709.5150 0.673335
2.60E-10 -0.684927
2045.180 -2.279066
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat
Prob.
0.25120.04750.65180.00080.01410.51350.50640.0417
14282.3424316.8222.0393622.4376522.117112.117940
TABLE 5
HETEROSKEDASTICITY TEST: WHITEF-statistic 8.102246 Prob. F (7, 12) 0.0009Obs*R-squared 16.50735 Prob. Chi-Square(7) 0.209Scaled explained SS 9.603122 Prob. Chi-Square(7) 0.2122Test Equation:Dependent Variable: RESID^2Method: Least SquaresDate: 04/9/14 Time: 14:39Sample: 1994 2013Included observations: 20
AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD PRICES 11
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
TABLE 6WEIGHTED LEAST SQUARE TEST
Dependent Variable: GOLD_PRICEMethod: Least SquaresDate: 04/11/14 Time: 20:23Sample: 1994 2013Included observations: 20Weighting series: (ABS(RESID))^(-0.5)Weight type: Inverse standard deviation (EViews default scaling)
Weighted Statistics
Uneighted StatisticsR-squaredAdjusted R-squaredS.E. of regressionDurbin-Watson stat
0.9620840.939966119.00991.950747
Mean dependent var S.D. dependent var Sum squared resid
670.1760485.7168169960.2
Variable
CUS_EXCHANGE_RATE^2
INFLATION_RATE^2CRUDE_OIL_PRICE^2
SHORT_TERM_INTEREST_RATE 2̂LONG_TERM_INTEREST_RATE^2
REAL_GDP^2SP500^2
R-squaredAdjusted R-squaredS.E. of regressionSum squared residLog likelihoodF-statisticProb(F-statistic)
Coefficient
3848.47113.70254
-9301.90021.87139
-16.53885-111.7561-0.000278-121.6887
0.9891620.98284056.6778838548.58
-104.0182156.46020.000000
Std. Error t-Statistic
826.4504 4.6566276.288245 2.1790721798.449 -5.1721791.544376 14.1619615.30461 -1.08064546.27543 -2.4150203.56E-05 -7.80162962.23501 -1.955309
Mean dependent varS.D. dependent varAkaike info criterionSchwarz criterionHannan-Quinn criter.Durbin-Watson statWeighted mean dop.
Prob.
0.00060.05000.00020.00000.30110.03260.00000.0742
794.9256597.467511.2018211.6001111.279571.886012567.0278
12 IJHPD VOL. 3 NO. 2 JULY - DEC 2014
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
TABLE 7HETEROSKEDASTICITY TEST: WHITE FOR WLS
F-statistic 0.874018 Prob. F(8,11) 0.5650Obs*R-squared 7.772442 Prob. Chi-Square(8) 0.4560Scaled explained SS 0.725686 Prob. Chi-Square(8) 0.9995
Test Equation:Dependent Variable: WGT_RESID^2Method: Least SquaresDate: 04/11/14 Time: 21:32Sample: 1994 2013Included observations: 20
Variable
CUS_EXCHANGE_RATE^2
INFLATION_RATE^2CRUDE_OIL_PRICE^2
SHORT_TERM_INTEREST_RATE 2̂LONG_TERM_INTEREST_RATE^2
REAL_GDP^2SP500^2
R-squaredAdjusted R-squaredS.E. of regressionSum squared residLog likelihoodF-statisticProb(F-statistic)
Coefficient
2246.753-15049.441.4244181765003.
-0.487207-21.5727781.690573.54E-11538.3493
0.388622-0.0560161463.56323562174-168.17300.8740180.565023
Std. Error t-Statistic
591.7512 3.79678710522.39 -1.4302302.129075 0.6690311757060. 1.0045210.523892 -0.92997464.74409 -0.33320176.18201 1.0723083.45E-11 1.024102424.5168 1.268146
Mean dependent varS.D. dependent varAkaike info criterionSchwarz criterionHannan-Quinn criter.Durbin-Watson stat
Prob.
0.00300.18040.51730.33670.37230.74520.30650.32780.2309
1112.5931424.21617.7173018.1653817.804772.046375
AN EMPIRICAL ANALYSIS OF FACTORS AFFECTING GOLD PRICES 13
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)
TA
BL
E 8
MU
LTIC
OLL
INE
AR
ITY
GO
LD_P
RIC
E
US_
EXC
HA
NG
E_R
ATE
INFL
ATIO
N_R
ATE
CR
UD
E_O
IL_P
RIC
E
SHO
RT_T
ERM
_INT
ERES
T_RA
T
LON
G_TE
RM_I
NTE
REST
_RAT
IREA
L_G
DP
SP50
0
GO
LD_
PRIC
E
1.00
0000
0.58
6697
-0.2
2411
6
0.90
8852
-0.7
3910
5
-0.8
3283
6
0.72
9926
0.37
7859
US_
EXC
HA
NG
E_R
ATE
0.58
6697
1.00
0000
-0.3
4432
6
0.60
7896
-0.7
0054
-0.7
5282
6
0.80
5677
-0.2
9881
3
INFL
ATIO
N_
RAT
E
-0.2
2411
6
-0.3
4432
6
1.00
0000
0.01
6805
0.40
7077
0.22
1364
-0.0
8897
2
-0.1
9414
3
CR
UD
E_O
IL_
PRIC
E
0.90
8852
0.60
7896
0.01
6805
1.00
0000
-0.6
9542
1
-0.8
7395
6
0.88
7450
0.22
0950
SHOR
T_TE
RM_
INTE
RES
T_R
ATE
-0.7
3910
5
-0.7
0054
0.40
7077
-0.6
9542
1
1.00
0000
0.89
7208
-0.7
2094
3
-0.1
5425
9
LON
G_TE
RM_
INTE
RES
T_R
ATE
-0.8
3283
6
-0.7
5282
6
0.22
1364
-0.8
7395
6
0.80
7208
1.00
0000
-0.9
1465
7
-0.1
3426
2
REA
L_G
DP
0.72
9926
0.80
5677
-0.0
8897
2
0.88
7450
-0.7
2094
3
-0.9
1465
7
1.00
0000
-0.1
1013
4
14 IJHPD VOL. 3 NO. 2 JULY - DEC 2014
Print to PDF without this message by purchasing novaPDF (http://www.novapdf.com/)