The Determinants of Consumer Price Index in Indonesia

pic THE DETERMINANTS OF CONSUMER PRICE INDEX IN INDONESIA Instructor DR. Moussa Larbani lively By Ali Faris(G0912449) Imala Hussain(G0822498) Ma Yue(G0918271) Mia Fathia(G0827756) Nurma Saleah(G0912298) Suthinee Suayngam(G0916798) Ulfah Hidayatun(G0815892) ECON 6030 ADVANCE QUANTITATIVE METHOD Term reputation Kulliyah of economic science and Management Sciences division of Business Administ proportionalityn 2009/2010 Abstract The some well known and commodiously quoted stinting indicator is the consumer cost indi puket finger (Consumer outlay Index).It represents an estimation of the change in worths of consumer right-hand(a)s and work. Gener altogethery, it represents a beat of our expenses on goods and services we utilise to meet our day-to-day needs. Severe bothers to the general saving privy be energised if the prices of consumer goods and services be abruptly changed. This paper attempts to examine the factors t assume deviate the Consumer hurt Index. We observe four variants, namely, m one and only(a)y try, unprocessed house servant product, participation rate, and shargon price.By utilizing quarterly data from 1996 to 2008, this study applies octuple degenerations method to regulate the best exercise and factors which back explain Consumer cost Index. The result indicates swinish internal product, recreate rate, and stock price portentous effect to consumer price index, whereas money try does not have significant effect. This study withal finds that the highest Adjusted R2 as goodness criteria of the model is derived when we include all the factors in the model.Hence, we can conclude that those factors have either strong or vague contri simplyion to consumer price index. Keyword Consumer set Index, 1. INTRODUCTION From the beginning of civilization, tribes, countries and nations have ever more than been looking for ship canal to attain prosperity and growth so as to improve the standard of living for their own quite a little. From the condemnations of Caesar to leaders of straightaway such(prenominal) as John F. Keneddy, things havent changed much. To attain prosperity one of the most important things is to maintain a healthy sparing.However on that point are many factors that threaten a healthy economy such as inflation, economic recessions and many otherwise factors. Despite all these threats and undeniable slumps and declines in economy, an economy can be monitored and as such Consumer charge Index is one of the most important economic indicators. Using consumer price index, the health of the economy can be in check and the several(prenominal)ize can take necessary pr chargetive measures otherwise not construe could lead to devastating effects in the form of high unemployment, bankruptcies, major(ip) financial losses etc.The cost-of-living index is a fixed-basket price index as it represents the price of a perpetual quantities basket of goods and services purchased by the conceive(a) consumer. cost-of-living index is one of the most frequently employstatistics for identifying periods of inflation or deflation. This isbecause large rises in consumer price index during a short period of time typically denote periods of inflation andlarge drops in CPI during a short period of time usuallymarkperiods ofdeflation. It is compiled by the Department of Labors Bureau of Labor Statistics.In order to get the final result for the CPI, wide researches of the prices of the included in the consumer basket goods and services are made. Then they are entered into a special computer program that makes the calculations. The importance of CPI is viewed in the fact that the estimations of other products, services and benefits are directly linked to the levels of the CPI. For example, if the CPI experiences an outgrowth in its note value, then the Social Securities benefits will rise as well. Other things that are directly linked to CPI include allowance Lease agre ements Union contracts Benefit statements and etc. Severe problems to the overall economy can be caused if the prices of consumer goods and services are abruptly changed. Most people associate the concept of CPI with inflation. An increase in the value of the CPI crockeds that an increase in inflation has been observed. When inflation increases the purchasing chroma of money is lost and people will change their spending habits as they meet their purchasing thresholds and producers will suffer and be forced to quash output. This can be readily tied to higher unemployment rates.The whole economy falls into a recession. The objective of this paper is to find a analogue obsession model that will accurately infer the consumer pricing index of Indonesia by using the following commutative variable quantitys, 1) money furnish, 2) Gross municipal product, 3) interest rates and 4) stock prices. In economics, money supply is the count amount of money available in an economy at a pa rticular point in time. There are several ways to define money, but standard measures usually include gold in circulation and demand deposits.The gross domestic product (GDP) or gross domestic income (GDI) is a basic measure of a countrys economic consummation and is the market value of all final goods and services made in spite of appearance the borders of a country in a year. It is a fundamental measurement of production and is very often positively cor colligate with the standard of living. An interest rate is the price a borrower pays for the use of money they do not own, for instance a small company index borrow from a bank to kick start their business, and the return a lender receives for deferring the use of funds, by lending it to the borrower.Interest rates are popularly expressed as a percentage rate over the period of one year. derivation Price in this paper is referred to as Stock check overet index which is based on a statistical compilation of the appoint price s of a number of representative stocks. We observe four variables namely, money supply, gross domestic product, interest rate, and stock price. By utilizing quarterly data from 1996 to 2008, this study applies nonuple relapsings method to find the best model and factors which can explain Consumer Price Index (See appendix 1 and 2). 2. METHODOLOGY 2. 1bivariate Pearson CorrelationPearson picis typically used to trace the strength of the analog kind amid twain quantitative variables. Often, these two variables are designated pic( prognosticator) and pic(outcome). Pearson pic has set that range from -1. 00 to +1. 00. The sign of picprovides reading about the worry of the relationship betwixt pic and pic. A positive correlation coefficient indicates that as lashings on pic increase, scads on pic besides tend to increase a nix correlation indicates that as establishs on pic increase, gobs on picneither increase nor decrease in a unidimensional manner.The absolute magnitu de of Pearson pic provides information about the strength of the analogue association between scores on pic and pic. For values of picclose to 0, on that point is no linear association between pic and pic. When pic= +1. 00, there is a double-dyed(a) positive linear association when pic= -1. 00, there is a perfect negatively charged linear association. In considerationediate values of piccorrespond to intermediate strength of the relationship (Warner, 2008). 2. 1. 1Assumption for Pearson pic (Warner, 2008)The assumptions that need to be met for Pearson pic to be an appropriate statistic to let on the relationship between a pair of variables are as follows 1. Each scores on pic should be independent of other pic scores (and each score on picshould be independent of other picscores). 2. Scores on two picand picshould be quantitative and normally distributed. 3. Scores on picshould be linearly related to scores onpic. 4. pic, picscores should have a bivariate normal statistical di stribution. 2. 1. calculation of Pearson pic (Warner, 2008) Formula to calculate Pearson picfrom the newfangled scores on picand picis as follows pic(2. 1) 2. 1. 3Correlation intercellular substance (Warner, 2008) A correlation hyaloplasm usually denoted by R it contains the correlations among all possible pairs of picvariables. The entire set of correlation in an R ground substance is as follow pic R = picpic Note several characteristics of this matrix. All the slash elements equal 1 (because the correlation of a variable with itself is, by definition, 1. 0).The matrix is symmetric because each element below the diagonal equals one comparable element above the diagonal. 2. 2Multiple Regressions Multiple Regression abridgment provides an comparability that predicts birthday suit score on a quantitative pic variable from stark scores on pic variables, with pic. The soothsayer or pic variables are usually to a fault quantitative, but it can also be a dichotomous variable (du mmy variable). Usually, degeneration analysis is used in non experimental research situations, in which the tec has manipulated none of the variables.In the absence of an experimental design, causal inferences cannot be made. However, researchers often have at least some of the predictor variables for retrogression analysis because they imagine that these might be causes of the outcome variable. If an pic variable that is theorized to be a cause of picfails to account for a significant amount of variate in the pic variable in the infantile fixation analysis, this outcome may weaken the researchers belief that the pic variable has a causal connection with pic.On the other hand, if a pic variable that is thought to be causal does uniquely predict a significant proportion of discrepancy in pic even when confounded variables or competing causal variables are statistically controlled, this outcome may be interpreted as consistent with the possibility of causality. (Warner, 2008) 1. The Multiple Regressions feigning Equation The raw score version of regression par with pic predictor variables is written as follows pic(2. 2) here picis the predicted score on the outcome (pic) variable, picis the intercept or constant term, picare regression coefficients, and picare predictor variables. The picregression coefficient represent partial slope. The pic slope represents the predicted change in pic for a one-unit increase inpic, controlling for pic(i. e. , controlling for all other predictor variables included in the regression analysis). The standard score version of a regression equation with pic predictors is represented as follows pic(2. ) where picis picscores on pic, picare beta coefficient that is used to predict The beta coefficients in the standard score version of the regression can be compared across variables to assess which of the predictor variables are more strongly related to the picoutcome variable when all the variables are represented in picscore form. Beta coefficient may be influenced by many types of artifacts such as unreliability of measurement and restricted range of scores in the try out. (Warner, 2008) 2. Model buildingThis paper use Stepwise regression model building to expose the least red-bloodeds regression in steps, either to forward selection disinclined elimination, or through standards stepwise regression. The coefficient of partial determination is the measure of the bare(a) contribution of each independent variable, given that other independent variables are in the model. 2. 2. 3Statistics Sum-of-squares foothold. Several regression statistics are computed as functions of the sums of-squares terms pic (2. 4) partitioning of variation.The regression equation is estimated such that the totality sum-of squares can be partitioned into components callable to regression and residuals SST = SSR+ SSE(2. 5) Coefficient of determination. The explanatory world power of the regression is summarized by its R-sq uared value, computed from the sums-of-squares terms as pic(2. 6) R2, also called the coefficient of determination, is often described as the proportion of variance accounted for, explained, or described by regression. It is important to celebrate in mind that a high R2 does not imply causation.The relative sizes of the sums-of-squares terms indicate how good the regression is in terms of fitting the standardisation data. If the regression is perfect, all residuals are zero, SSE is zero, and R2 is 1. If the regression is a total failure, the sum-of-squares of residuals equals the total sum-of-squares, no variance is accounted for by regression, and R2 is zero. Adjusted R2. The R2 value for a regression can be made arbitrarily high simply by including more and more predictors in the model. The adjusted R2 is one of several statistics that attempts to revivify for this artificial increase in accuracy.The adjusted R2 is given by pic(2. 7) n = sample size (e. g. , number of years of data in calibration period) p = number of predictors in the model, not counting the constant term As shown by the equation, R2 with hat is lower than R2 if the model has more than one predictor. Adding predictors has the effect of increasing the difference between R2 with hat and R2. Adjusted R2 is also useful in comparing among models. ANOVA parry and definition of concoct squared terms. The sums-of-squares terms and related statistics are often summarized in an compendium of Variance (ANOVA) dodge pic Source= source of variationSS= sum-of-squares term df= degrees of freedom for SS term MS= stiff squared terms The mean squared terms are the sums-of-squares terms Standard erroneousness of the estimate. The residual mean square (MSE) is the sample estimate of the variance of the regression residuals. The population value of the error term is sometimes written as ? e2 while the sample estimate is given by se2 = MSE(2. 8) where MSE has been delimit previously. The square root of the residual mean square is called the root-mean-square error (RMSE), or the standard error of the estimate. pic(2. 9) The subscript c is attached (RMSEc) in (4. ) to distinguish the RMSE derived from calibration from the root-mean-square error derived by cross-validation (see later). F ratio or overall F. Recall that the explanatory power of a regression is given by the regression R2, which is computed from sums-of-squares terms. The F-ratio, or overall F, which is computed from the mean squared terms in the ANOVA table, estimates the statistical substance of the regression equation. The F-ratio is given by pic(2. 10) The advantage of the F- ratio over R2 is that the F- ratio takes into account the degrees of freedom, which depend on the sample size and the number of predictors in the model.A model can have a high R2 and still not be statistically significant if the sample size is not large compared with the number of predictors in the model. The F- ratio incorporates sample size and number of predictors in an assessment of significance of the relationship. The significance of the F- ratio is obtained by referring to a table of the F distribution, using degrees of freedom df1,df2, where df1 and df2 are the degrees of freedom for the regression mean square and residual mean square from the ANOVA table.How to reject or accept F-test (for overall significance) HO ? 1 = ? 2 HA ? 1 and ? 2 not both zero ? = . 05 Decision Reject Ho if the f-stat falls in the rejection area (p values ? = . 05) pic T-test. The T-test shows if there is a linear relationship between the variable xi and y. The test statistic pic(2. 11) How to reject or accept T-test (for individualist significance) HO ? 1 = 0 HA ? 1 ? 0 ? = . 05 Decision Reject Ho if the test statistic for each variable falls in the rejection region (p values . 05) picConfidence interval for estimated coefficients. If the regression assumptions on the residuals are satisfied, including the normality assumption, t hen the sampling distribution of an estimated regression coefficient is normal with a variance proportional to the residual mean square (MSE). The variance of the estimator also depends on the variances and covariances of the predictors. The idea is best illustrated for the case of unbiased linear regression, for which the variance of the regression coefficient is given by pic(2. 12)Where Se2 is the residual mean square, xi is the value of the predictor in year xi with hat is the mean of the predictor, and the summation is over the n years in the calibration period. The 100 (1 ? ?) % confidence interval is pic, where t? /2 is obtained from s t distribution with n-2 degrees of freedom. For more than one predictor, the confidence intervals for regression can be computed similarly, but the equation is more complicated. The equation for the variances and covariances of estimated coefficients is expressed in matrix terms by pic(2. 13) where X is the time series matrix of predictors.This equation returns a matrix, with the variances of the parameters a colossal the diagonal, and the covariances as the off-diagonal elements (Weisberg 1985, p. 44). The appropriate degrees of freedom of the t distribution is df = n ? K ? 1, where K is the number of predictors in the model, and n is the sample size. Multicolinearity The predictors in a regression model are often called the independent variables, but this term does not imply that the predictors are themselves independent statistically from one another. In fact, for natural systems, the predictors can be highly intercorrelated. Multicolinearity is a term reserved to describe the case when the intercorrelation of predictor variables is high. It has been noted that the variance of the estimated regression coefficients depends on the intercorrelation of predictors. Haan (2002) concisely summarizes the effects of multicolinearity on the regression model. Multicolinearity does not invalidate the regression model in the sense that the prophetic value of the equation may still be good as long as the prediction are based on combinations of predictors within the kindred multivariate space used to calibrate the equation.But there are several negative effects of multicolinearity. First, the variance of the regression coefficients can be idealistic so much that the individual coefficients are not statistically significant even though the overall regression equation is strong and the predictive ability good. Second, the relative magnitudes and even the signs of the coefficients may defy interpretation. For example, the regression fish on a tree-ring index in a multivariate regression equation to predict precipitation might be negative even though the tree-ring index by itself is positively correlated with precipitation.Third, the values of the individual regression coefficients may change radically with the removal or add-on of a predictor variable in the equation. In fact, the sign of the coefficient migh t even switch. Signs of multicolinearity. Signs of multicolinearity include 1) high correlation between pairs of predictor variables, 2) regression coefficients whose signs or magnitudes do not make good physical sense, 3) statistically non-significant regression coefficients on important predictors, and 4) extreme sensitivity of sign or magnitude of regression coefficients to insertion or deletion of a predictor variable.Variance Inflation Factor (VIF). The Variance Inflation Factor (VIF) is a statistic that can be used to identify multicolinearity in a matrix of predictor variables. Variance Inflation refers here to the mentioned effect of multicolinearity on the variance of estimated regression coefficients. Multicolinearity depends not just on the bivariate correlations between pairs of predictors, but on the multivariate predictability of any one predictor from the other predictors. Accordingly, the VIF is based on the multiple coefficient of determination in regression of each predictor in multivariate linear regression on all the other predictors pic(2. 14) where Ri2 is the multiple coefficient of determination in a regression of the ith predictor on all other predictors, and i VIF is the variance inflation factor associated with the ith predictor. Note that if the ith predictor is independent of the other predictors, the variance inflation factor is one, while if the ith predictor can be almost perfectly predicted from the other predictors, the variance inflation factor approaches infinity. In that case the variance of the estimated regression coefficients is unbounded.Multicolinearity is said to be a problem when the variance inflation factors of one or more predictors becomes large. How large it appears to be a subjective judgement. According to Haan (2002), some researchers use a VIF of 5 and others use a VIF of 10 as a critical threshold. These VIF values correspond, respectively, to Ri2 values of 0. 80 and 0. 90. Some compute the medium VIF for al l predictors and adjudge that an average considerably larger than one indicates multicolinearity (Haan, 2002).At any rate, it is important to keep in mind that multicolinearity requires strong intercorrelation of predictors, not just non-zero intercorrelation. The VIF is closely related to a statistic call the tolerance, which is 1/VIF. Some statistics packages answer for the VIF and some report the tolerance (Haan 2002). 3. MODEL SPESIFICATION AND DATA SOURCE Based on the opening review in the previous section, we build the following specification to ictus the determinants of money supply in Indonesia CPI = ? 0 + ? 1M1 + ? 2GDP + ? 3IR + ? 4SP + ? The variables are defined as followed 1.Money supply (M1) is M0 (physical currency) and demand deposits, which are checking accounts. This is used as a measurement for economists trying to quantify the amount of money in circulation. The M1 is a very liquid measure of the money supply, as it contains cash and assets that can quickly b e converted to currency. 2. Gross Domestic Product (GDP) is the income of individuals or nations after adjusting for inflation. 3. Consumer price index (CPI) is an index number measuring the average price of consumer goods and services purchased by households. 4. Interest rate (IR) is a gift paid on borrowed capital. . Share Price (SP) is the price of one share of stock. This paper uses quarterly data, from quarter 1 of 1996 to quarter 2 of 2008 that is taken from International Financial Statistic. We also use SPSS software to regress the model above. 4. VALUATION 4. 1Model Estimation We will present the result of data analysis using Multiple Regression Analysis. Multiple Regression analysis provides an equation that predicts raw score on a quantitative pic variable from raw scores on pic variables, withpic. The best model is indicated by the highest Adjusted R2 and the last(a) standard errors.In this study, consumer price indexes (CPI) were predicted from the following variables money supply (M1), gross domestic product (GDP), interest rate (IR) and share price (SP). The sample sizepicis 50. 4. 2Bivariate correlation In this part, we will observe the strength of the linear relationship between each independent variable and CPI. Table 1. Correlations CPI M1 GDP IR SP 1996Q1 11. 716111 19. 30 4. 771904 3. 788341 3. 853983 0. 065642 1996Q2 11. 766373 19. 4 4. 819983 3. 822246 3. 858643 0. 036398 1996Q3 11. 827298 19. 17 4. 717570 3. 877778 3. 863081 -0. 014697 1996Q4 11. 879324 19. 16 4. 810590 3. 909344 3. 872063 -0. 037281 1997Q1 11. 889998 18. 98 4. 934683 3. 905124 3. 897606 -0. 007518 1997Q2 11. 914423 18. 72 4. 941414 3. 921527 3. 906252 -0. 015275 1997Q3 12. 002958 23. 38 4. 78997 4. 036802 3. 924765 -0. 112037 1997Q4 12. 039144 26. 19 4. 477901 4. 111444 3. 959830 -0. 151614 1998Q1 12. 262335 26. 33 4. 624532 4. 270861 4. 140733 -0. 130127 1998Q2 12. 314070 32. 16 4. 495154 4. 363053 4. 309088 -0. 053965 1998Q3 12. 484700 34. 93 4. 308177 4. 530508 4. 491942 -0. 038566 1998Q4 12. 457244 35. 20 4. 294247 4. 12515 4. 538626 0. 026111 1999Q1 12. 510708 34. 11 4. 396215 4. 536868 4. 585091 0. 048223 1999Q2 12. 512071 30. 34 4. 767910 4. 478028 4. 578437 0. 100409 1999Q3 12. 533785 24. 52 4. 754038 4. 455257 4. 555728 0. 100470 1999Q4 12. 525806 21. 68 4. 830264 4. 422381 4. 555029 0. 132648 2000Q1 12. 689215 19. 58 4. 798267 4. 536245 4. 79349 0. 043104 2000Q2 12. 725801 18. 46 4. 615507 4. 572988 4. 589384 0. 016396 2000Q3 12. 796032 17. 98 4. 534614 4. 630870 4. 611431 -0. 019440 2000Q4 12. 817033 17. 80 4. 436443 4. 654596 4. 639514 -0. 015082 2001Q1 12. 894097 17. 85 4. 423641 4. 715383 4. 668689 -0. 046693 2001Q2 12. 957670 18. 26 4. 396349 4. 769620 4. 695093 -0. 74527 2001Q3 12. 980581 18. 88 4. 453272 4. 786409 4. 731538 -0. 054871 2001Q4 12. 967675 19. 20 4. 357638 4. 787355 4. 758569 -0. 028785 2002Q1 13. 014972 19. 32 4. 495629 4. 812124 4. 804455 -0. 007668 2002Q2 13. 038967 19. 18 4. 670443 4 . 813862 4. 813371 -0. 000492 2002Q3 13. 083051 18. 87 4. 499660 4. 860963 4. 830240 -0. 030724 2002Q4 13. 67842 18. 42 4. 383610 4. 856585 4. 856372 -0. 000213 2003Q1 13. 119451 18. 20 4. 382903 4. 894774 4. 879052 -0. 015721 2003Q2 13. 127729 17. 68 4. 568618 4. 880785 4. 881073 0. 000288 2003Q3 13. 168067 16. 44 4. 706932 4. 890676 4. 889544 -0. 001132 2003Q4 13. 145558 15. 43 4. 867750 4. 851847 4. 910358 0. 058511 2004Q1 13. 193018 14. 0 5. 023394 4. 869902 4. 926710 0. 056808 2004Q2 13. 243557 14. 28 5. 023446 4. 905169 4. 946239 0. 041070 2004Q3 13. 296856 13. 88 5. 055704 4. 940406 4. 956855 0. 016449 2004Q4 13. 303815 13. 54 5. 255827 4. 925389 4. 972241 0. 046852 2005Q1 13. 357168 13. 36 5. 375579 4. 954380 5. 001198 0. 046817 2005Q2 13. 415743 13. 29 5. 03708 4. 996403 5. 019906 0. 023503 2005Q3 13. 477237 13. 78 5. 410051 5. 046527 5. 037628 -0. 008900 2005Q4 13. 539065 15. 78 5. 387751 5. 110080 5. 135998 0. 025918 2006Q1 13. 570606 16. 34 5. 539352 5. 12 4611 5. 157502 0. 032891 2006Q2 13. 608447 16. 23 5. 624725 5. 145281 5. 164111 0. 018831 2006Q3 13. 676882 16. 00 5. 78345 5. 191494 5. 176234 -0. 015260 2006Q4 13. 679898 15. 35 5. 839146 5. 174745 5. 194761 0. 020015 2007Q1 13. 732362 14. 70 5. 885796 5. 206364 5. 219177 0. 012812 2007Q2 13. 777640 14. 08 6. 039466 5. 223014 5. 222613 -0. 000401 2007Q3 13. 848229 13. 56 6. 144445 5. 264208 5. 239273 -0. 024935 2007Q4 13. 855779 13. 11 6. 305412 5. 52354 5. 259836 0. 007483 2008Q1 13. 930695 12. 94 6. 292750 5. 309960 5. 292817 -0. 017143 2008Q2 14. 023264 12. 95 6. 172412 5. 392000 5. 199684 -0. 192316 0. 175227 From the table above, we found the sum square value of error is 0. 175. Predict Consumer Price index (CPI) for a quarter in which the logarithmic of GDP is 12. 89 logarithmic of Interest Rate is 17. 5 and logarithmic of Share Price is 4. 42 LCPI = -4. 927 + 0. 769 (LGDP) + 0. 007 (LIR) 0. 090 (LSP) = -4. 927 + 0. 769 (12. 89) + 0. 007 (17. 52) 0. 090 (4 . 42) = 4. 72 Confidence interval for the mean LCPI value pic pic pic pic foretelling interval for the mean LCPI value pic pic pic pic CONCLUSION We have employed multiple regression analysis method, which involve five variables which are expected to poignant money supply. They are consumer price index, interest rate, stock price, GDP, and money supply M1. The data are selected from Indonesia international financial statistics.In the recent years Indonesia has been successfully controlling its money supply to get stability in economic circumstances. From the study we found out that there is strong relationship between consumer price index CPIand GDP. When the Gross Domestic Product GDP increases, it will also increase consumer price index as these two have linear relationship. Also there is strong correlation between money supply and consumer price index, which means that mean of CPI increase when money supply increases. Addition to this there is positive correlation between sto ck price and CPI, when stock price increase it tend to increase CPI.However there is negative correlation between interest rate and CPI, when interest rate increases, CPI decreases. From our finding it shows that R-square is 96 percent, which means it is a good model to describe the relation between CPI and other variables we use in the study. REFERENCES Lawrence S. Meyers, Glenn Gamst, and A. J. Guarino. (2006). apply Multivariate Research Design and Interpretation. Thousand Oaks, London, and New Delhi Sage Publications. Miles, Jeremy and Mark Shevlin. (2001). Applying Regression & Correlation A Guide for Students and Researchers. London Sage Publications. Warner, R. M. (2008).Applied Statistics From Bivariate Through Multivariate Techniques. Los Angeles, London, New Delhi, Singapore SAGE Publications. Watson, Collin J. and et al. (1993). Statistics for Management and Economics 5th Edition. Massachusetts Allyn and Bacon. http//www. investopedia. com http//www. stock-market-investo rs. com http//www. wikipedia. org Appendix 1. Variables Data M1 Stock Price CPI INTEREST RATE GDP 1996Q1 53162. 00 118. 14 47. 8 19. 30 122530. 00 1996Q2 56448. 00 123. 96 47. 40 19. 24 128846. 00 1996Q3 59684. 00 111. 90 47. 61 19. 17 136940. 00 1996Q4 64089. 00 122. 80 48. 04 19. 16 144253. 00 1997Q1 63565. 00 139. 03 49. 28 18. 98 145801. 00 1997Q2 69950. 00 139. 97 49. 71 18. 72 149406. 00 1997Q3 66258. 00 118. 99 50. 64 23. 8 163237. 00 1997Q4 78343. 00 88. 05 52. 45 26. 19 169252. 00 1998Q1 98270. 30 101. 96 62. 85 26. 33 211575. 00 1998Q2 109480. 00 89. 58 74. 37 32. 16 222809. 00 1998Q3 102563. 00 74. 30 89. 29 34. 93 264263. 00 1998Q4 101197. 00 73. 28 93. 56 35. 20 257106. 00 1999Q1 105705. 00 81. 14 98. 01 34. 11 271226. 0 1999Q2 105964. 00 117. 67 97. 36 30. 34 271596. 00 1999Q3 118124. 00 116. 05 95. 18 24. 52 277558. 00 1999Q4 124633. 00 125. 24 95. 11 21. 68 275352. 00 2000Q1 124663. 00 121. 30 97. 45 19. 58 324232. 00 2000Q2 133832. 00 101. 04 98. 43 18. 46 336314. 00 2000Q3 135430. 00 93. 19 100. 63 17. 98 360783. 00 2000Q4 162186. 0 84. 47 103. 49 17. 80 368440. 00 2001Q1 148375. 00 83. 40 106. 56 17. 85 397956. 00 2001Q2 160142. 00 81. 15 109. 41 18. 26 424077. 00 2001Q3 164237. 00 85. 91 113. 47 18. 88 433905. 00 2001Q4 177731. 00 78. 07 116. 58 19. 20 428341. 00 2002Q1 166173. 00 89. 62 122. 05 19. 32 449087. 00 2002Q2 174017. 00 106. 5 123. 15 19. 18 459993. 00 2002Q3 181791. 00 89. 99 125. 24 18. 87 480725. 00 2002Q4 191939. 00 80. 13 128. 56 18. 42 473469. 00 2003Q1 181239. 00 80. 07 131. 51 18. 20 498546. 00 2003Q2 195219. 00 96. 41 131. 77 17. 68 502690. 00 2003Q3 207587. 00 110. 71 132. 89 16. 44 523382. 00 2003Q4 223799. 00 130. 03 135. 9 15. 43 511733. 00 2004Q1 219087. 00 151. 93 137. 93 14. 80 536605. 00 2004Q2 226147. 00 151. 93 140. 65 14. 28 564422. 00 2004Q3 234676. 00 156. 92 142. 15 13. 88 595321. 00 2004Q4 245946. 00 191. 68 144. 35 13. 54 599478. 00 2005Q1 244003. 00 216. 07 14 8. 59 13. 36 632331. 00 2005Q2 261814. 00 222. 23 151. 40 13. 9 670476. 00 2005Q3 267762. 00 223. 64 154. 10 13. 78 713000. 00 2005Q4 271166. 00 218. 71 170. 03 15. 78 758475. 00 2006Q1 270425. 00 254. 51 173. 73 16. 34 782779. 00 2006Q2 303803. 00 277. 20 174. 88 16. 23 812968. 00 2006Q3 323885. 00 292. 47 177. 02 16. 00 870551. 00 2006Q4 347013. 00 343. 49 180. 33 15. 35 873181. 0 2007Q1 331736. 00 359. 89 184. 78 14. 70 920214. 00 2007Q2 371768. 00 419. 67 185. 42 14. 08 962838. 00 2007Q3 400075. 00 466. 12 188. 53 13. 56 1033260. 00 2007Q4 450055. 00 547. 53 192. 45 13. 11 1041090. 00 2008Q1 409768. 00 540. 64 198. 90 12. 94 1122080. 00 2008Q2 453093. 00 479. 34 181. 22 12. 95 1230910. 00 Appendix 2. Lag of Variable Data lm1 lsp lcpi lgdp ir 1996Q1 10. 881099 4. 771904 3. 853983 11. 716111 19. 30 1996Q2 10. 941075 4. 819983 3. 858643 11. 766373 19. 24 1996Q3 10. 996819 4. 717570 3. 863081 11. 827298 19. 17 1996Q4 11. 068028 4. 810590 3. 872063 11. 879324 19. 16 1997Q1 11. 059818 4. 934683 3. 897606 11. 889998 18. 98 1997Q2 11. 55536 4. 941414 3. 906252 11. 914423 18. 72 1997Q3 11. 101311 4. 778997 3. 924765 12. 002958 23. 38 1997Q4 11. 268852 4. 477901 3. 959830 12. 039144 26. 19 1998Q1 11. 495477 4. 624532 4. 140733 12. 262335 26. 33 1998Q2 11. 603497 4. 495154 4. 309088 12. 314070 32. 16 1998Q3 11. 538233 4. 308177 4. 491942 12. 484700 34. 93 1998Q4 11. 524824 4. 294247 4. 538626 12. 57244 35. 20 1999Q1 11. 568407 4. 396215 4. 585091 12. 510708 34. 11 1999Q2 11. 570855 4. 767910 4. 578437 12. 512071 30. 34 1999Q3 11. 679490 4. 754038 4. 555728 12. 533785 24. 52 1999Q4 11. 733129 4. 830264 4. 555029 12. 525806 21. 68 2000Q1 11. 733369 4. 798267 4. 579349 12. 689215 19. 58 2000Q2 11. 804341 4. 615507 4. 589384 12. 725801 18. 46 2000Q3 11. 16210 4. 534614 4. 611431 12. 796032 17. 98 2000Q4 11. 996499 4. 436443 4. 639514 12. 817033 17. 80 2001Q1 11. 907498 4. 423641 4. 668689 12. 894097 17. 85 2001Q2 11. 983816 4. 396349 4 . 695093 12. 957670 18. 26 2001Q3 12. 009066 4. 453272 4. 731538 12. 980581 18. 88 2001Q4 12. 088026 4. 357638 4. 758569 12. 967675 19. 20 2002Q1 12. 020785 4. 495629 4. 804455 13. 14972 19. 32 2002Q2 12. 066908 4. 670443 4. 813371 13. 038967 19. 18 2002Q3 12. 110613 4. 499660 4. 830240 13. 083051 18. 87 2002Q4 12. 164933 4. 383610 4. 856372 13. 067842 18. 42 2003Q1 12. 107572 4. 382903 4. 879052 13. 119451 18. 20 2003Q2 12. 181877 4. 568618 4. 881073 13. 127729 17. 68 2003Q3 12. 243306 4. 706932 4. 889544 13. 168067 16. 44 2003Q4 12. 18504 4. 867750 4. 910358 13. 145558 15. 43 2004Q1 12. 297224 5. 023394 4. 926710 13. 193018 14. 80 2004Q2 12. 328941 5. 023446 4. 946239 13. 243557 14. 28 2004Q3 12. 365961 5. 055704 4. 956855 13. 296856 13. 88 2004Q4 12. 412867 5. 255827 4. 972241 13. 303815 13. 54 2005Q1 12. 404936 5. 375579 5. 001198 13. 357168 13. 36 2005Q2 12. 475390 5. 403708 5. 19906 13. 415743 13. 29 2005Q3 12. 497854 5. 410051 5. 037628 13. 477237 13. 78 200 5Q4 12. 510486 5. 387751 5. 135998 13. 539065 15. 78 2006Q1 12. 507750 5. 539352 5. 157502 13. 570606 16. 34 2006Q2 12. 624135 5. 624725 5. 164111 13. 608447 16. 23 2006Q3 12. 688144 5. 678345 5. 176234 13. 676882 16. 00 2006Q4 12. 757118 5. 839146 5. 194761 13. 679898 15. 5 2007Q1 12. 712095 5. 885796 5. 219177 13. 732362 14. 70 2007Q2 12. 826025 6. 039466 5. 222613 13. 777640 14. 08 2007Q3 12. 899407 6. 144445 5. 239273 13. 848229 13. 56 2007Q4 13. 017125 6. 305412 5. 259836 13. 855779 13. 11 2008Q1 12. 923346 6. 292750 5. 292817 13. 930695 12. 94 2008Q2 13. 023853 6. 172412 5. 199684 14. 023264 12. 95 pic pic pic