Predicting Life Expectancy

Only available on StudyMode
  • Download(s) : 175
  • Published : December 12, 2012
Open Document
Text Preview
The quest to prolong our youth continues today, though not through lengthy field explorations, but through improvements in health, nutrition, and medicine. A healthy diet, regular exercise, and vaccinations can greatly improve an individual’s life expectancy, while an outbreak of disease, malnutrition, and social unrest can drastically lower an individual’s life expectancy. But how are life expectancies affected on a national level?

While these factors are central to living longer, they alone cannot be the only facets. The social and economic conditions of each country will undoubtedly affect its citizens, their lifestyles and decisions. Citizens of wealthier countries have access to modern medicine and medical facilities, the leisure to exercise, and meticulous regulation of sanitation and drinking water. Their life expectancies, therefore, naturally should be higher than those of less developed countries. However, this is not always the case. According to the World Health Organization (WHO), the United State of America ranked 24th overall in terms of life expectancy among all countries in the year 2000. Japan, Australia, France, Sweden, Spain, Italy, Greece, and Switzerland, all ranked above the more developed United States.This paper uses mostly regression to estimate life expectancy for countries around the world from social and governmental factors provided by The World Health Organization. First, I describe the data and methods used in the analysis. Next, I provided results from my analysis, and finally, I conclude with my findings. .

It is assumed that our chosen economic and social variables exert an observable and significant influence on life expectancy at birth of all nations. The relationship between life expectancy and these variables is assumed to be linear and subject to random error.

. As much as we would have loved to collect data across all nations of the world, this data was not readily available. Although finding most country data was simple, at times it downright tedious. Several countries do not release the statistics of factors that affect their economic and social conditions. When performing a multiple regression analysis, any category that contains missing data will be left out of the data set. Thus, data from all countries of the world could not be gathered and therefore were excluded from the regression. Although this would have ensured the most complete regression model, we assume our sample of 121 countries is a good reflection of this overall world population, and that variables significant in our model will also apply to all nations of the world. Once again due to the inconsistencies with the data, the values we collected for our variables were inconsistent. Although most of the data was recorded in the year 1995, this was not always the case. It is far too expensive for data to be collected in every nation during every year. Therefore, several of our variables were collected during different year, and several collected over a span of several years. However, never was data collected more than two years before or after 1995. It is assumed that extreme fluctuations in the social and economic conditions did not occur during these years, and thus this data set is appropriate.Life expectancy and the input variables in the regression analysis are provided by The World Health Organization at The World Health Organization is the leading authority of health related matters for countries in the United Nations.

The following data, by country was obtained from The World Health Organization. Only countries with all of the following variables were used in the analysis so that the data was complete:

* Life expectancy at birth (Y) – This is the response variable, and it measures how long a person born in a country is expected to live. * Average price of cigarettes (X1) – This input variable is the price as of...
tracking img