Revenue vs. Education in the U.S. and the United Kingdom

2332 WordsJul 8, 201810 Pages
4.0. Analysis and Results In this chapter, statistical results of the revenue vs. education in The USA and in The UK will be comparatively illustrated. The time period chosen lies between 2008 and 2013 (immediately after the effects of the financial crisis started to appear, and up until today); firstly, data will be presented via bar charts and statistical information, and will continue with a regression for each country which will illustrate the qualitative parameters of the chosen model, and will establish the amount of influence between the “education level” and “annual income” series. The fixed model (or the Ordinary Least Square approach) is the most suitable model for our datasets, according to the result of the Hausman test. 4.1.…show more content…
If we are to look again at the bar charts from the beginning of this chapter, we can also observe the fact that the increase of the wage, in time, is very slow, and in the case of 2008-2009, a visible decrease had even been registered. 4.1.3. OLS Regression results Dependent Variable: MEDIAN_ANNUAL_WAGE Method: Least Squares Date: 02/25/14 Time: 12:32 Sample: 1 30 Included observations: 29 Variable Coefficient Std. Error t-Statistic Prob. C (1) 25545.74*** 4151.713 6.153061 0.0000 C (2) 16162.59*** 1666.441 9.698867 0.0000 R-squared 0.776985 Mean dependent var 58985.59 Adjusted R-squared 0.768725 S.D. dependent var 25899.07 S.E. of regression 12455.14 Akaike info criterion 21.76413 Sum squared resid 4.19E+09 Schwarz criterion 21.85842 Log likelihood -313.5798 F-statistic 94.06801 Durbin-Watson stat 1.147248 Prob(F-statistic) 0.000000 Figure 5: OLS regression results for The USA dataset The regression results show that the R-square value is 0.776, meaning that the model could be explained by the independent value in a proportion of 77%. This is a significant value, which reveals that 77 percent of the overall factors that influence the income are related to the education level of an individual. The p-value of the model (or probability value) is less than 0.05, which means that the model is statistically significant; the t-statistics also reveals a significant value of 9.69,