The GDP per capita plays an important role in determining a country’s logistics performance. There is a moderately strong relationship between all logistics performance indicators and GDP. This finding implies that the lower the GDP score of a country, the more likely it is to have a low logistics performance score and vice versa.
After fitting the best possible regression models, logistics performance scores versus GDP per capita, some extreme values are found in the regression procedure.
These unusual findings are depicted in Table 1a and 1b (See the attachment). Table 1a of unusual residuals lists all observations that have studentized residuals greater than 3 in absolute value. Table 1b of unusual residuals lists all observations that have studentized residuals less than 3 but greater than 2 in absolute value.
Studentized residuals measure how many standard deviations each observed value deviates from a model fitted using all of the data except that observation.
Turkay,
first of all one should always remember, as you point out in your question, that the relationship between the two issues is moderate. There is, in my opinion, also an issue of heterogeneity in definition, and measurement accuracy that might induce unwanted heterogeneity in the data. Furthermore, the logistic demand is induced by GDP, and it can safely be regard as derived from it. But, the most relevant issue for the data you consider is the specific countries that you consider as peculiar. They are either very large (India China) or very small and landlocked (Luxembourg) or islands (Haiti, Madagascar). I am not all that impressed that the relationship seems problematici in these cases.
Hope it helps
Regards
Edoardo
- Badges
- Science topic
- Similar topics
- Books