Current Issues in Economic Integration - Statistics Project Example

Data Analysis-STATA DATA ANALYSIS In the analysis of the impact of gross domestic product, GDP Growth and Gross Capital Formation on foreign direct investment in BRIC countries, regression models will be used. The models capture the changes of variables as a result of tie changes. The regression mode thus becomes; FDI is the dependent variable, abbreviated as Y GDP independent variable, abbreviated as X1 GDP growth is an independent variable, abbreviated as X2 GCF is an independent variable, abbreviated as X3 Thus the model is simplified to; Summary statistics The data has 160 observations that include 40 sets of data from the four BRIC countries. There are 3 predictor variables against one outcome variable. The predictor variables are X1, X2, X3 and the outcome variable Y. The mean show the averaged value of the observations. On the other hand, the standard deviation indicates the extent through which a value has deviated from the mean. Variable Y shows that highest deviation at 26408.94. The variable X1, X2, and X3 have lesser deviations with 2896.749, 5.558 and 11.676 respectively. The period covered by the data is 40 years from 1973 to 2012. Hausman test: Fixed or Random effects In order to determine the model that will be used in the analysis, it is important to conduct the Hausman test to ascertain the characteristics of the data and whether it poses random effects or fixed effects. The difference are shown by the Prob>chi2 value. At a significance level of 5%, the Prob>chi2 value should be less than 0.05. The table of results below shows the Hausman test for the data. From the Hausman test result table above, it can be deduced that the variables do not pose random effects but rather fixed effects. This is because the results pose a Prob>chi2 = 0.0001 is less than the required value of >0.05. Therefore, henceforth, the discussion will involve treat the fixed effects model. Consequently, any hypothesis on random effects model will be nullified. Regression output tables for the model The model was regressed at a confidence level of 95%. The variables X1, X2, and X3 were regressed against the dependent variable Y. Interpretation of the table Variable X1 (GDP) At a significance level of 95% confidence level, the variable X1 (GDP) is highly significant in the model with a p-value P>|t| = 0.0000. This implies that the variable X1 is of significant influence in the model that predicts the dependent variable. Thus, the probability is high that the coefficient is not zero, thus predicting the significance of a variable in the model and a strong indicator of a positive relationship between the dependent variable Y and independent variable X1. The variable X1 has a t-value of 10.30. This implies that the variable is significant to the model though it has extremities that make the t-statistic relatively high. The standard error is important in creating confidence intervals. The standard error of the GDP is 0.6008 that is lower than the corresponding coefficient of the same variable. A standard error that is lower that the coefficient implies that there is high chance the declarative hypothesis (H=0) is not valid, thus the model is significant. The variable GDP has a coefficient of 6.1909. The coefficient shows the magnitude through which the variable X1 affects the independent variable Y. In this case, the dependent variable, foreign direct investment changes by 6.1909 due to a unit change in the GDP of a country that is member of BRIC. Variable X2 (GDP growth) The variable has a P>|t|=0.126. At a significant level of 5%, the variable is not significant to the model. Thus, its impact on the model is not optimal. Moreover, at a confidence level of 95%, the t-statistic of the variable is 1.54. The recommended t-statistic value or 95% confidence level is 1.96. Therefore, the variable is not significant at 95% significance level. The standard error of the variable is relatively high at 337.3629. This implies that it would open a wide confidence interval thus reducing the accuracy of the results. However, since the standard error is lower than the corresponding coefficient, it may influence the hypothesis of the model. A coefficient of 519.118 indicates that a unit change in GDP growth increases the foreign direct investment by 519.118. X2 shows a positive relationship with the independent variable. The implication of this is that if an economy grows and the GDP increases by a single unit, then the amount of foreign direct investment would shift positively by 519.118. X3 (Gross Capital Formation) The P>|t|=0.096, implying that the variable is not significant at 95% confidence level. Moreover, a t-statistic of 1.67 is insufficient at 95 % confidence interval. At least the value should be 1.96 at 95%confidence interval. The standard error is also higher in relation to the coefficient. The variable has a standard error of 201.158 against a coefficient of 336.6516. The direct implication is that the null hypothesis applies to the model (H≠0). The coefficient of the Gross capital formation is 336.6516. It follows that a unit change in the Gross capital formation accounts for a change of 336.6516 in Foreign Direct Investment. The model has an R-square of 0.5981. The R-square shows the level of variance in the dependent variable that is explained by the variables in the model. The extent through which the model explains the dependent variable is represented by the r-square. Significance of the overall model The significance of the whole model is of interest to the study. The significance is determined by the relationship between the f-calculated and F-critical using the following rule of thumb: If F-critical is greater than F-calculated: the model is not significant that is, (F-critical> F-calculated) If F-calculated is greater than F- critical: the model is significant, that is, (F-critical< F-calculated) From the ANOVA table, the value of F calculated is 697432140 whereas the value of F-critical is 37.95. Therefore, it can be concluded that the overall model is statistically significant since F- calculated is greater than F-critical (697432140>37.95). The overall p-value of the model is Prob > F = 0.0000 meaning that the model is significant at 5% significance level. Empirical Analysis of the data For the analysis of the determinants of foreign direct investments, the fixed effects mode will be employed as proven by the Hausman test. This is because it will enable to make viable conclusions regarding the impact of all the variables on the predictor variable. In this regard, individual entities of the variables with unique characteristics and vary over time will be regressed against the dependent variable to ascertain its full impact on the predictor variable. The fixed effects model removes all biased characteristics of data such a time invariant aspects that can hinder the discovery of the net effect of the predictor (Torres-Reyna n.d). The fixed effects model is also suitable in reducing the impact of correlation on the outcome. Correlation of variables corrupts the data. In addition, since each variable is unique and has its unique characteristics, the constant and the sets error term should not correlate with other values from other variables. The equation for the fixed effects model becomes: Yit= β1Xit+ αi+ uit Where αi (i=1….n) is the unique intercept for every entry (n entity-specific intercepts). Yit is the dependent variable (DV) in which i = entity and t = time. Xit denotes one independent variable (IV), β1 is the coefficient for the independent variable uit is the error term So, in the case of FDI against the variables; GDP, GDP growth and Gross capital formation, regressions involving the dependent variables on each independent variable will be carried out to assess the impact of each individual variable on the dependent variable. Gross domestic product and foreign direct investment As shown in the statistics above, foreign direct investment is determined by a number of factors that also determine the extent of the investments. Factors such as GDP, Economic growth and increase in capital formation capabilities of countries have significantly influence FDI. The BRICs countries have a high capability of attracting foreign investments due to several factors such as raw material availability, labour abundance and low cost of doing business. From the table, GDP is the most prominent factor that influences the foreign direct investment. The coefficient of 6.744985 and a standard error 0.5366346 implies that GDP of a country accurately influences the amount of foreign direct investments that will flow into the country. This is so because a change in the domestic product of a country within the BRICs category produces an increase in the GDP by 6.190916 units. From the dataset, the country that has the highest amount of FDI is the one with a high gross domestic product. China records the highest amount of FDI than any other BRIC member. The impact of GDP on foreign direct investment can be shown by regressing the GDP against FDI holding other factors constant. Suppose the Ordinary Least Squares is used to estimate the impact of GDP on foreign investment, the table of results would be as follows. The rho value of 0.5133 mean that 51.33% of all the variance arises from the differences across the panel. The rho value is also referred as the interclass correlation. Sigma _u is the standard deviation of residuals that are in group u. Sigma_e is the standard deviation of the overall error terms. Since the result depict a stable positive relationship between foreign direct investment and GDP, then it can be concluded that changes in GDP will always impact on the FDI status of a BRIC country. The positive coefficient of the constant predicts that foreign direct investment can be influenced by 1739.583 by variables that are not in the model, that is, the autonomous value of FDI when not under the influence of other variables. In a nutshell, the results of the analysis proves that members of the BRIC are reliant on the GDP to attract foreign investment. Brazil, for example had a high GDP that attracted investor thus recording surpluses in 2004 (Edwards 2009: 350).The positive relationship imply that GDP and FDI implies that the countries that have a higher GDP have more potential to attract more foreign investment. The mutuality of GDP and FDI is that while they depend on each other, they also influence each other positively. This is because a country with stable GDP wins investors’ confidence in doing business. GDP growth and foreign direct investment The effect of GDP growth on Foreign Direct Investment is also positive. Suppose we regress the dependent variable Y (FDI), against the variable X2 (GDP growth, we get the following table of results. The table shows that, holding other factors constant, an increase in GDP is likely to increase the FDI of BRIC countries. Change in GDP becomes significant at 5% when other factors are held constant with a p-values of 0.038 and a t-value of 2.10. The results strongly suggest that there is a linear relationship between the variables thus an increase in one variable must affect the other variable in the same direction. The relationship between the square of the sigma_u and sigma_e gives the rho. It predicts the extent of variance that is explained by the difference observed across the panels. In this case, 17.565% of all the variance is caused by these differences. The difference between the coefficient and the standard error is large enough to ensure that declarative hypothesis (H=0) is rejected. By having such a difference, it means that the null hypothesis would be accepted since H≠0). When other variables are held constant, foreign direct investment changes by 10506.82. The change that is represented by the coefficient of the constant arises from other exogenous variables and other variables that may be correlated to the variables under consideration. The empirical explanation for the results is that in an increase in a country’s GDP leads to an increase in the foreign direct investments. As stated above, when a country’s worth is growing, investors gain confidence about it and thus put more resources into the country. Growth in GDP thus improves a country’s image, creditworthiness and thus paints a good image on the rating of doing business. From the data, the countries under BRIC pose a unanimous trend that paint a vivid picture on the impact of GDP growth on FDI. Returns from investments improve when a country’s portfolio grows and this value for the doing business improves. The analysis proves that the growth of GDP in BRIC countries is the reason of in increased FDI in these countries. The inference is reached due to the fact that BRIC countries are emerging economies characterised by fast growing economies and thus target for foreign direct investments. FDI and Gross capital formation Investors prefer countries where the expected returns are high. Capital formation involved addition in the value of assets owned in a country. In this regard, FDI tends to flow to area with high returns thus the gross capital formation and foreign direct investment have a positive relationship. This relationship can be shown in the following table of a regression of Y against X3 as shown in the table. The coefficient for X3 (Gross capital formation) is positive. This implies that a unit increase or decrease in capital formation changes the FDI by 1325.792. The relationship implies that in BRIC countries, the value of capital assets and the expected changes within the time of investment is determinant of foreign investment. In addition, the standard deviation is lower if all the other determinants are held constant. If we consider capital formation as the only factor in the model, the model is significant at 95% confidence level with t-value of 5.71 and p-value of 0.000. This show the magnitude in which the variable affects foreign direct investment. The coefficient of the constant (-16269.04) predicts the changes that would occur to the independent variable due to other factors not considered in the model. The change is autonomous or external to the model of FDI and Capital formation. Capital formation is critical as it shows gradual improvement in the net worth of a business. Thus, it is logical enough to conclude that firms would possibly go to area where the returns are optimal, and also where their business can expand. In summary, the estimation of the relationship between Foreign Direct Investment and its determinants is important for the BRIC countries that heavily rely on investments. Countries like China and India have experience massive growth due to foreign direct investments and are heavily depend on it (Andreosso-OCallaghan, Zolin 2013: 41). From the analysis above, all the determinants posed a positive relationship when regressed against foreign direct investment. This shows that critical factors such as gross domestic product, capital formation and also GDP growth are critical for a country to attain high levels of foreign investments. Bibliography Andreosso-OCallaghan B., & Zolin B. 2013. Current Issues in Economic Integration: Can Asia Inspire the West?. Ashgate Publishing. Edwards, S. 2009. Capital controls and capital flows in emerging economies: Policies, practices, and consequences. Chicago: University of Chicago Press. Torres-Reyna, O. n.d. Panel Data Analysis Fixed & Random Effects. Data and Statistical Services. Princeton University. Available from http://dss.princeton.edu/training/Panel101.pdf Walsh & Yu. 2010. Determinants of Foreign Direct Investment: A Sectorial and Institutional Approach. IMF working Paper. International Monetary Fund. Read More