Influence of Gross Domestic Product and Income Distribution on Gross National Income - Essay Example

Modelling the Statistics of Economy Influence of Gross Domestic Product and Income distribution on Gross National Income Introduction This essay is an econometric analysis of the various economical statistics in terms of specific model that best describes them. The major statistics of economy as described in NationMaster website (http://www.NationMaster.com ) include Gross Domestic Product (GDP), Purchasing Power Parity (PPP), distribution of income, distribution of consumption, gross national income, GDP and Real Growth rate, Human development index, Budget, Expenditures, technology index, informal economy, debt and Economic freedom of contries (http://www.NationMaster.com/graph/eco_eco_fre-economy-economic-freedom). These statistics are inter - dependent on one another and are influenced by one another. The influence may be in positive direction or negative direction. Their influence can be best described by appropriate models that characterise their relationship. Chosen Econometric issue The Gross National Income for countrie is an economic index reflecting the Gross Domestic product (GDP) and other economic statistics. While GDP is the net value of all producers along with the taxes, Gross national income is sum of GDP and all income from abroad. Bothe GDP and GNI talk about the economic status of a country. It is a measure of the economic development of a country involving the nations expenditure, income and national output. the GDP is an estimate of the kiving standard of a country as the per capita measure of GDP measures the average income per person of the nation. The Purchasing Power Parity (PPP) of GDP is the measure of the GDP in terms of international currency. The GDP (PPP) per capita is the variation divided among the population. The income inequality or income distribution has been great concern for economist in the recent past. This dimension highly influences the political and social dimension of a country. Both developing and developed countris tend to concentrate on equality in income distibution (ILO, 2004). Isabel and Matthew (2011) have discussed income inequality in terms of market exchange rates, national history, PPP exchange rates, across different regions of the country and the takeaway. They also discuss about the income distribution pattern among poor, women and children. The widely used measure of income inequality is in terms of Gini index, which is a coeffecient measured in terms of Lorenz curve. If each person has same income, then, the Gini coeffecient is Zero (0). A Gini coeffecient of One (1) conveys that income is confined to only one person in the population. This project intends to concentrate on these aspects (GDP, PPP, Income distribution and Gross National income). The econometric model that has been planned for implementing this project is the regression model. A causal model is planned one dependent variable and two independent variable. The regression is expected to be positive among all the variables. Data Set Used The data for this project has been acquired from the Economy>Statistics link of the NationMaster.com. this data repository has a wide collection of all data related to economic situations . The statistics are listed country wise for ease of use and understanding. Three sets of economic statistics are used in this project, which include – “Income distribution > Richest 10% by country", (http://www.NationMaster.com/graph/eco_inc_dis_ric_10-economy-income-distribution-richest-10), "GDP > PPP by country", (http://www.NationMaster.com/graph/eco_gdp_ppp-economy-gdp-ppp) and "Gross National Income by country", Retrieved from (http://www.NationMaster.com/graph/eco_gro_nat_inc-economy-gross-national-income). The data set “GDP >PPP”, presents the most recent GDP measures of countries. The PPP is represented in terms of millions of international dollars. A snapshot of the data source is presented in Fig.1. Fig.1. The GDP – PPP datasource ((http://www.NationMaster.com/graph/eco_gdp_ppp-economy-gdp-ppp) For this project 53 observations from this dataset has been chosen as seen in the log file of STATA. The summary statictics of this variable showin the mean of all data, the standard deviation, maximum value, minimum value is shown in Table. 1. Table.1 Summary Statistics of the data – GDP > PPP summarize var1 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- var1 | 53 9.35e+11 1.91e+12 3.09e+09 1.16e+13 The second dataset Income distribution presents the list of richest countries in terms og income distribution (http://www.NationMaster.com/graph/eco_inc_dis_ric_10-economy-income-distribution-richest-10). 53 nations that are present in the earlier list were chosen. A snapshot of this data source is shown in Fig.2. and the summary statistics for this dataset is tabulated in Table 2. Fig. 2. The data source of Income distribution (http://www.NationMaster.com/graph/eco_inc_dis_ric_10-economy-income-distribution-richest-10). Table 2. The summary statistics for Income distribution summarize var2 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- var2 | 53 29.23585 7.477739 18.2 48 The third data set used in this project is the Gross National Income data taken from NationMaster.com. 53 nations that are present in bothe the earlier lists, were chosen. This data set is presented in Fig.3. and the corresponding summary statistcis in Table.3. Fig. 3. The Gross National Income data (http://www.NationMaster.com/graph/eco_gro_nat_inc-economy-gross-national-income). Table 3. Summary statistcis for Gross National Income summarize numvar3 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- numvar3 | 53 5.40e+11 1.48e+12 3.09e+09 9.78e+12 The Econometric Model The chosen model is a causal model with one dependent variable and two independent variables. The independent variables influence the dependent variable. The causal model could be based on individual causal effect of the independent variables or the average causal effect of the independent variables (Neyman, 1990). For this project, individual causal effect of the independent variableis considered. The chosen dependent variable is the “Gross National Income” and the independent variables are “Income distribution” and “GDP – PPP”. The causal model can be framed as in Fig.4. Fig.4. The Causal model for this project in terms of dependent and independent variables. Econometric Method used The econometric method used for this project is the causal regression method. Statistical regression provides the explanation of one variable in terms of another variable. It enables comparison of one variable with another. The right side variables (independent variables) influence the left side (dependent variable). The nature of relation between the variables can be assessed by a scatter plot and a model equation (Niles, Robert, 2006). This model equation is in terms of coefficients of regression along with a constant. Regression analysis aids in future prediction. Simple regression is used to find the relation between one dependent variable and one independent variable (Moore, 2006). Multiple regression gives the regression of more than one independent variables on the dependent variable. Regression analysis can be used for testing hypotheses in terms of the regression coefficient, standard error, squared value of regression coefficient, degree of freedom, ‘F’ value for the specified degree of freedom. For analysis, initially a null hypothesis Ho and an alternate hypothesis Ha are defined. Based on the outcome of regression analysis, the null hypothesis Ho will be accepted or rejected. If the regression coefficients are not significant, then the null hypothesis Ho will be rejected. This implies that the alternate hypothesis is accepted. For this project, the hull hypothesis and the alternate hypothesis are chosen as, Ho : GDP and income distribution DO NOT influence the Gross National Income. Ha : GDP and income distribution influence the Gross National Income. After obtaining the regression outcome, the regression results are subjected to a diagnostic test like Box- Cox, Forecasting, fitting with least squares, Ramsey Reset, etc..). For this project work, the fitting of residuals is considered as the diagnostic test. This diagnostic test provides the distribution of the residuals on a line. The level of dispersion of the residuals has to be less for best fit. In other words, the residuals must lie closer to the line of fit. This project has been implemented in STATA Release (12) the statistical package for analysing data in terms of all inferential and descriptive statistics. The STATA package is capable of handling logitudinal data, panel data, time series data and survey data. The analyses can be in terms of multivariate analysis and survival analysis. STATA also performs Structural Equation Modelling (SEM). The type of regression analysis available in STATA many - linear regression, constrained linear regression, censored regression, truncated regression, box cox regression, quantile regression and multivariate regression under multiple equation models (www.stata.com) . For this project, the linear regression is chosen as it suffices the requirement of regressing two independent variables on one dependent variable without any constraints. The commond link or menu for implementing this linear regression is Statistics > linear models and related > linear regression. In the option, the dependent and independent variables are entered. The variables need to be in numeric format. Apart from this, the confidence interval has to be specified, the default confidence interval (CI) is 95%. This implies a tailed test for the hypothesis testing. The ‘F’ value considered for decision on the hypothesis depends on the confidence interval. If Fmodel < F distribution , the regression model is not significant and the null hypothesis is rejected. Results The regression results for this project is shown in Table. 4. Which has the linear regression of GDP – PPP (var1) and Income distribution (var2) on the dependent variable Gross National Income (numvar3). The result shows the regression coefficients, standard error, ‘t’ value and ‘F’ value for aa confidence interval of 95%, number of observations, degrees of freedom, root mean square error, squared regression coefficient and adjusted value of squared regression coefficient. Table.4. Regression Outcome regress numvar3 var1 var2 Source | SS df MS Number of obs = 53 -------------+------------------------------ F( 2, 50) = 90.89 Model | 8.9010e+25 2 4.4505e+25 Prob > F = 0.0000 Residual | 2.4482e+25 50 4.8964e+23 R-squared = 0.7843 -------------+------------------------------ Adj R-squared = 0.7757 Total | 1.1349e+26 52 2.1825e+24 Root MSE = 7.0e+11 ------------------------------------------------------------------------------ numvar3 | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- var1 | .6812321 .0506899 13.44 0.000 .5794184 .7830458 var2 | -1.81e+10 1.30e+10 -1.40 0.169 -4.42e+10 7.94e+09 _cons | 4.33e+11 3.93e+11 1.10 0.276 -3.57e+11 1.22e+12 ------------------------------------------------------------------------------ The regression line that fits relation between the chosen variables is, numvar3 = 4.33e+11 + 0 .6812321 var1 - 1.81e+10 var2. where GDP – PPP is var1, Income distribution is var2 and Gross National Income is numvar3. From the above regression outcome, standard error due to var1 is less (0.05068), but the same for var2 is very high. Also, the standard F value for the given degree of freedom F( 2, 50) = 90.89 is high compared to the F value from model. So Fmodel < F standard, which implies that the null hypothesis has to be rejected. And the alternate hypothesis is accepted. Ho : GDP and income distribution DO NOT influence the Gross National Income – is rejected. Ha : GDP and income distribution influence the Gross National Income – is accepted. Based on the regression coefficient, the influence of income distribution is not significant but the influence of gross domestic product – purchasing power parity is highly significant. The outcome of the regression diagnostic is the graph shown in Fig. 5. Fig.5. The Fitting of residuals for the regression analysis. From the figure, it can be seen that most of the residuals are close to each other and only a few of them are dispersed. This shows that the performed regression analysis is valid and that the project outcome is reliable. Conclusion This project has presented the regression analysis of three different econometric variables. The variables are macro economic variables like Gross Domestic Product – Purchasing Power Parity, Income distribution and Gross National income. The data from valid statistics database has been collected. A sample spaace of 53 observations have been tabulated under these three variables. The econometric model chosen for analysing these variables is a causal model that has been tested with regression analysis. Linear multiple regression has been conducted with two independent variables (Gross Domestic Product – Purchasing Power Parity, Income distribution) on the dependent variable (Gross National income). It has been found out that the null hypothesis does not hold as the F statistic and the regression coefficients do not support them. Hence, the chosen alternate hypothesis Ha : GDP and income distribution influence the Gross National Income – is accepted. Also a regression diagnostic has been performed on the outcome of the regression residuals and the model has been validated and is true. The major conclusions are that the Gross Domestic Product – Purchasing Power Parity highly influences the Gross National Income but the Income distribution does not influence the Gross National Income. References "Economic freedom by country", The Heritage Foundation. Retrieved from http://www.NationMaster.com/graph/eco_eco_fre-economy-economic-freedom Accessed 21 Jan. 2014. Economy Statistics , Nation master.com http://www.NationMaster.com Accessed 21 Jan. 2014. "GDP > PPP by country", World Bank. 2005. World Development Indicators 2005.. Retrieved from http://www.NationMaster.com/graph/eco_gdp_ppp-economy-gdp-ppp Accessed 21 Jan. 2014. "Gross National Income by country", . Retrieved from http://www.NationMaster.com/graph/eco_gro_nat_inc-economy-gross-national-income Accessed 21 Jan. 2014. ILO (2004). A Fair Globalization: Creating Opportunities for All. Report of the World Commission on the Social Consequences of Globalization. Geneva: International Labour Organisation. “Income distribution > Richest 10% by country", World Bank. 2002. World Development Indicators 2002. CD-ROM. Washington, DC. Retrieved from http://www.NationMaster.com/graph/eco_inc_dis_ric_10-economy-income-distribution-richest-10 Accessed 21 Jan. 2014. Isabel Ortiz, Matthew Cummins (2011), “Global Inequality :Beyond the Bottom Billion - A Rapid Review of Income Distribution in 141 Countries”, Policy, Advocacy and Knowledge Management, Division of Policy and Practice, UNICEF, 3 UN Plaza, New York, NY 10017 Moore, David (2006). “Basic Practice of Statistics.” WH Freeman Company. pp 90–114. Neyman, J. (1990). On the application of probability theory to agricultural experiments. Essay on principles. Section 9. Statistical Science, 5, 465-472. Niles, Robert, 2006. "Robert Niles Journalism Help: Statistics Every Writer Should Know," RobertNiles.com (http://www.robertniles.com/stats/. ) Accessed 21 Jan. 2014. “STATA QUICK REFERENCEANDINDEX”, RELEASE 12 www.stata.com Accessed 21 Jan. 2014. Appendices 1. Log files in STATA 2. The data set used Si. No COUNTRY GDP RICH GROSS NATIONAL INCOME 1 US 11628083000000.00  30.5   9780000000000.00  2 CHINA 7123712000000.00  30.4   1130000000000.00  3 JAPAN 3774086000000.00  21.7   4520000000000.00  4 INDIA 3362960000000.00  33.5   477000000000.00  5 GERMANY 2325828000000.00  23.7   1940000000000.00  6 UK 1832252000000.00  27.7   1480000000000.00  7 FRANCE 1744352000000.00  25.1   1380000000000.00  8 ITALY 1621372000000.00  21.8   1120000000000.00  9 BRAZIL 1482859000000.00  48   529000000000.00  10 RUSSIA 1408603000000.00  38.7   253000000000.00  11 SPAIN 1046249000000.00  25.2   588000000000.00  12 MEXICO 1014514000000.00  41.7   550000000000.00  13 CANADA 993079000000.00  23.8   682000000000.00  15 INDONESIA 779719000000.00  26.7   145000000000.00  16 AUSTRALIA 605942000000.00  25.4   386000000000.00  17 TURKEY 552990000000.00  32.3   167000000000.00  18 NETHER 520918000000.00  25.1   390000000000.00  19 THAI 510268000000.00  32.4   118000000000.00  20 SOUTH AFRICA 510102000000.00  45.9   122000000000.00  22 POLAND 499549000000.00  24.7   164000000000.00  24 PHILIPPINES 378225000000.00  36.6   80844900000.00  25 PAKISTAN 336050000000.00  27.6   60047300000.00  27 BELGIUM 322645000000.00  23   245000000000.00  28 COLOMBIA 322582000000.00  46.1   81551500000.00  29 UKRAINE 303280000000.00  23.2   35185000000.00 ` 31 SWEDEN 265048000000.00  20.1   226000000000.00  32 BANGLA 263434000000.00  28.6   48616900000.00  33 AUSTRIA 261106000000.00  22.5   195000000000.00  34 SWITZERLAND 247602000000.00  25.2   277000000000.00  35 MALAYSIA 246036000000.00  38.4   79326600000.00  36 GREECE 243130000000.00  25.3   121000000000.00  37 VIETNAM 222172000000.00  29.9   32761600000.00  39 ALGERIA 210657000000.00  26.8   51028000000.00  40 PORTUGAL 205450000000.00  28.4   109000000000.00  41 CZECH 197367000000.00  22.4   54309900000.00  42 CHILE 183286000000.00  45.6   70619200000.00  43 ROMANIA 182343000000.00  25 38616800000.00  44 NORWAY 175435000000.00  21.8   161000000000.00  45 DENMARK 172569000000.00  20.5   164000000000.00  46 HUNGARY 167584000000.00  20.5   49161600000.00  47 ISRAEL 163703000000.00  28.3   107000000000.00  48 IRELAND 160993000000.00  27.4   87735600000.00  50 NIGERIA 155571000000.00  40.8   37132000000.00  51 FINLAND 155489000000.00  21.6   123000000000.00  52 PERU 155388000000.00  35.4 52209300000.00  53 MOROCCO 127228000000.00  30.9   34681400000.00  55 KAZAKHSTAN 112091000000.00  26.3   20078200000.00  57 SRI LANKA 81144000000.00  28 16411300000.00  58 SLOVAKIA 78262000000.00  18.2   20307200000.00  59 TUNISIA 76508000000.00  31.8   19984500000.00  61 GHANA 48747000000.00  30.1   5749130000.00  62 GEORGIA 3094980000.00  27.9   3094980000.00  63 ZAMBIA 3328150000.00  41   3328150000.00  Read More