Descriptive statistics of GDP and median age

The economists have a growing interest in Social And Political Environment. Moreover, the data given is collected for 125 countries in the year 2000. On the other hand, the regression has been carried out on variables like gdp, urbanization rate, median age of populations and political right index

As per the question, descriptive statistics of the gdp and medage has been given as:

The median age of the population is mainly 26 years of age depicting that the youth is the major part of the population whereas real GDP depicts that 8454.97 is the average real GDP experienced in 125 countries. Moreover, the standard deviation of the real GDP is 8736.38. This shows that there is a lot deviation from the mean depicting that the real GDP data is varied drastically among countries (). The same can be said for median age of population amongst the countries that is highly deviated from the mean. The median that depicts that equal number across the two halves of the sample has 5141 as gdp and 24 medage showing that the median is lower than mean. In addition, this can also be interpreted that the developing nation’s growth process has been at low end based on low gdp per capita (Boulhol, Serres and Molnar 2008).

Conversely, there is high positive skewness in gdpc data than in medage. The gdpc data shows skewness of 1.49 and that of medage is 0.38. However, the data is not normally distributed as mean, median and mode are not equal. Moreover, kurtosis also explains the shape of the distribution. The data of gdpc has positive kutosis showing that the data has leptokurtic shape and medage has negative kurtosis depicting platykurtic shape that is flatter in nature than the normally distributed curve (mesokurtic) (Etienne, Devogele and McArdle 2014).

As per the diagram of real GDP (dollars) with median age of population, it can be seen that the real GDP is concentrated to the working population from working age. As a result, real GDP results to be scattered and slightly grows with the working population’s growth that is 15 years of age to 42 years of age.As per the diagram of real GDP (dollars) with urbanization rate, it can be seen that the real GDP is concentrated to the rate from to 1 and widely amongst the 125 countries. As a result, real GDP results to be scattered and is increasing at an increasing rate as the gdp is rising amongst the countries (Kitov 2006).

Regression analysis of real GDP and independent variables

èlgdpc = 5.338817 + 0.36638*plr + 0.06634*medage + 2.10549*ur

The regression analysis has been carried using Excel (illustrated in Appendix 1). However, the output shows that all the independent variables has positive yet direct impact on real GDP whether it is political right index, median age of population or urbanization rate.

Theslope coefficient of median age of population portrays that any change in median age of population will lead to 6.634% change in the real GDP per capita.

Theslope coefficient of urbanization rate represents that any change in urbanization rate will lead to 210.549% change in the real GDP per capita. This change can be highly effective.

As per the output by Excel (Appendix 1), the predicted and actual values of Australia and Japan in 2000 can be given in the table below.

As per the table made based on the Appendix 1, it can be seen that the actual values of Real GDP of Australia and Japan is more than the predicted values of real GDP in 2000. This states that the developed economies have performed well in the year 2000 ().

The estimated model can be given as:

èlgdpc = 5.338817 + 0.36638*plr + 0.06634*medage + 2.10549*ur

The β₁ slope coefficient of plr represents that any change in political right index will lead to 36.638% change in the real GDP per capita. This change is significant on the dependable variable real GDP because p value = 0.029 which is less than p = 0.05 at 95% confidence level.

The overall significance of the model is based on the F statistics calculated using excel in appendix 1. The F statistics shows that not all values are jointly equal to zero (Frost 2015). In this case degrees of freedom are df₁ = 3 and df₂ = 121. Moreover, the table value of F statistics at 0.05 level is 2.7 and the calculated F statistics is 166.215 which states F_table < F_calculated at 0.05 level stating and proving that the independent variable coefficients jointly are not equal to zero. Hence, this elaborates that the results are statistically even significant because p value is even less than 0.05 (Frost 2016). The same can be depicted through residual plots.

The significance of the model is even depicted through R square and adjusted R square and in this model it is 80.47% and 79.9886% respectively. This states that the model is best fit to be applied as there are less variations in the independent variables (Seltman 2012). Moreover, it also helps in analyzing that relationship between the model and response variable is strong and significant ().

Significance of political right index, median age of population, and urbanization rate on real GDP

The analysis of the data is on economic data for Thailand and highlights the GDP, CPI, Exchange rate and interest rate.

ècpi_t = 95.18304 + 0*gdp_t – 1.059971*exrate_t – 0.227026*ir_t

The regression analysis has been carried using Excel (illustrated in Appendix 2). However, the output shows that all the independent variables has null yet negative effect on CPI whether it is GDP, exchange rate or interest rate respectively.

As per the significance of the independent variables at p value =0.05 at 95% confidence level, GDP (p = 0.00 < p =0.05), exchange rate (p = 0.00 < p =0.05) are significant as there p values is lesser than 0.05 confidence level. However, the same cannot be said for interest rate (p = 0.07 > p = 0.05) because p value is greater than 0.05 confidence level (Frost 2016).

Re-estimation of the model by single lag

Cpi_t-1 = β₀ + β₁gdp_t-1 + β₂exrate_t-1 + β₃ir_t-1 + µ_t-1

èCpi_t-1 = 94.30503 + 0*gdp_t-1 – 0.995515*exrate_t-1 – 0.262327*ir_t-1

The regression analysis has been carried using Excel (illustrated in Appendix 3). However, the output shows that all the independent variables even while using single lag has null yet negative effect on CPI whether it is GDP, exchange rate or interest rate respectively.

The long run impact of the interest rate and exchange rate can be further analyzed such as the monetary policy is affected by interest rate. On the other hand, exchange rate is affected by the appreciation/depreciation of currency value (Bartram and Bodnar 2012).

The results depict that interest rate and exchange rate have negative effect on inflation such that any affect in exchange rate will lead to negative 105.99% changes in CPI and any affect in interest rate will lead to negative 22.7026% changes in CPI. This can be further elaborated that exchange rate has a stronger negative effect then the interest rate on inflation index (CPI).

The lagged values in the single lag model Cpi_t-1 = 94.30503 + 0*gdp_t-1 – 0.995515*exrate_t-1 – 0.262327*ir_t-1 are significant based on the F statistics calculated using excel in appendix 3. The F statistics shows that not all values are jointly equal to zero. In this case degrees of freedom are df₁ = 3 and df₂ = 71. Moreover, the table value of F statistics at 0.05 level is 2.76 and the calculated F statistics is 139.71 which states F_table < F_calculated at 0.05 level stating and proving that the independent variable coefficients jointly are not equal to zero (Frost 2016). Hence, this elaborates that the results are statistically even significant because p value is even less than 0.05. The same can be depicted through residual plots.

References

Bartram, S.M. and Bodnar, G.M., 2012. Crossing the lines: The conditional relation between exchange rate exposure and stock returns in emerging and developed markets. Journal of International Money and Finance, 31(4), pp.766-792. Available at: https://mpra.ub.uni-muenchen.de/14018/1/MPRA_paper_14018.pdf [Accessed 12 May 2017].

Boulhol, H., Serres, A. and Molnar, M. 2008. The Contribution of Economic Geography to GDP per Capita. [online] oecd.org. Available at: https://www.oecd.org/eco/42506177.pdf [Accessed 12 May 2017].

Etienne, L., Devogele, T. and McArdle, G., 2014. Oriented spatial box plot, a new pattern for points clusters. International Journal of Business Intelligence and Data Mining, 9(3), pp.233-253. Available at: https://eprints.maynoothuniversity.ie/6084/1/oriented_spatial_box_plot_etienne_devogele_mcardle.pdf [Accessed 12 May 2017].

Frost, J. 2015. What Is the F-test of Overall Significance in Regression Analysis? | Minitab. [online] Blog.minitab.com. Available at: https://blog.minitab.com/blog/adventures-in-statistics-2/what-is-the-f-test-of-overall-significance-in-regression-analysis [Accessed 12 May 2017].

Frost, J. 2016. Understanding Analysis of Variance (ANOVA) and the F-test | Minitab. [online] Blog.minitab.com. Available at: https://blog.minitab.com/blog/adventures-in-statistics-2/understanding-analysis-of-variance-anova-and-the-f-test [Accessed 12 May 2017].

Frost, J. 2016. Understanding Hypothesis Tests: Significance Levels (Alpha) and P values in Statistics. [online] Blog.minitab.com. Available at: https://blog.minitab.com/blog/adventures-in-statistics-2/understanding-hypothesis-tests%3A-significance-levels-alpha-and-p-values-in-statistics [Accessed 12 May 2017].

Kitov, I. 2006. Real GDP per capita in developed countries. [online] arxiv.org. Available at: https://arxiv.org/ftp/arxiv/papers/0811/0811.0889.pdf [Accessed 12 May 2017].

Seltman, H., 2012. Experimental Design and Analysis Available from https://www. stat. cmu. edu/~ hseltman/309/Book. Book. pdf.