Sunday, February 08, 2009

Paper for the May gradudate conference in social economy is coming along.

I started to scan with my eyes divergences from the perfect population pyramid. I have some doubts now about my interpretation and criteria for booms, after finding, in fact, the world population has some age bands starting in 2000 where the older band exceeds the younger band in size. This is my first clue of a boom or divergence from the perfect pyramid. This rule is, if an older age band exceeds a younger age band, we have identified a boom. I have scanned the available world pyramids for both genders combined but not separately and copied and pasted the full years data for all occurrences of the divergent pattern. I then tried to construct a population pyramid with Excel. I needed to code one gender's population with negative numbers to get the Excel chart to look something like a population pyramid. I also scanned with my eyes Afghanistan and next is Albania.

So my method is to scan with my eyes. The other choice of method is to read the tables into R and run a check to see if a table decreases continuously from the 0-4 age band to the 80+ age band. Actually the 80+ age band is giving me some false positives in the rule above. I will do all the countries by eye and then run the R code to check. Or better would be to next design the R code to find all the divergences and not have to scan all the rows of countries. I have so far scanned only the World and Afghanistan and reached about the thousandth row of thirty three thousand rows of data. So here automating will save me some time.

Also getting the data from each table into R gives me more statistical power than simply identifying one divergent pattern. I can run many pattern recognitions once the data is cleanly into R or other statistical software.

Here is my abstract for this paper.

The world over all has no baby boom. There are always more babies and fewer seniors when one looks at the whole world. We would like to examine if baby booms exist in any and each country. If this examination is possible we will then move further. The study will survey the age pyramid structures, the typical retirement age and industries, that the majority of the population work within, for all the world's nations, if data can be located. If these measures are possible attempts are made at assessing the potential knowledge losses in these majority industries as the widest parts of the age pyramid retire and take their workplace operational knowledge with them potentially crippling these industries and nations. General comments are attempted concerning each nation studied.

No comments: