This project is intended to help you to improve your understanding of the point
estimation and the interval estimation of the population mean of the indicator selected
from the Penn World Tables 9.0 and to build statistical analysis’ skills.
As we have discussed since the start of this class, collecting and analyzing data are not
easy tasks. In particular, collecting data on a population of interest requires a lot of
resources including time and financial means.
The sample allows you to estimate the parameters of the population using the sample
data. How close is the point estimate to the population parameter cannot be determined?
However, the interval estimation can provide additional information on the closeness
between the estimate and the population parameter.
For this project, you will use secondary data to estimate the mean of the population of
interest (point and interval estimation) and compare such mean for two regions (a poor
and a rich region). Regions include North America, Australia, New Zealand, and Western
European countries (1), African countries (2), Latin American countries (3), Eastern
European countries and Russia (4), and Asian countries (5).
Structure This project needs to be structured as follows:
A succinct description of the issue you are researching on the population of interest, the
source of data (provided), the sampling method to be used for the analysis, etc. For each
region, consider the reported data as the sample and select just one year. Data are
reported only for a few countries.
Implementation of the method:
- Description of how the sample was collected (size of population must be the total
number of countries on earth, and the size of the sample must be the number of
countries in the selected region with available data.
- Estimate the value of the population mean (point estimate) and the standard
deviation using sample data for each of the two selected regions.- Construct the confidence intervals for the population mean for each region using
the confidence level of your choice (preferably between 90% and 99%). If the
sample size of your region is 30 or more (30 countries of more), assume that your
S = σ and use Zα/2 to construct the confidence interval. But, if the sample size of
your region is less than 30 countries, use tα/2 to construct the confidence interval.
- Show the results on the graph.
- Next, determine the sample size needed to make the confidence interval more
precise given your selected margin of error and the confidence level
- Construct the new confidence interval using the new margin of error (must be the
same for the two regions).
Describe the meaning of your results and how they can be used for policy
Maximum: three pages (not included the appendix for graphs).
Sources of data
Penn World Tables 9.0 (indicators highlighted in yellow).
PLEASE SEE THE ATTACH