Wednesday, February 24, 2010

Life Expectancy: Does my Insurance Company know about This?

In the previous post about life expectancy, we looked at some data of the last 20 years and extrapolated linearly. It seemed as if men and women would finally live equally long somewhere in the next decennium.

While thinking and reading about the way life expectancy is calculated, it struck me that the calculation is not fair. I started out describing the way the calculation is done and why I think it is wrong. During this, I found out that the Wikipedia page about this topic already contained the answer:

It is important to note that this statistic is usually based on past mortality experience, and assumes that the same age-specific mortality rates will continue into the future. Thus such life expectancy figures are not generally appropriate for calculating how long any given individual of a particular age is expected to live. But they are a useful statistic to summarize the current health status of a population.


Basically, in the calculation, one assumes that when you are born does not influence the probability with which you will die (at a certain age). This is obviously false.

The article further describes that models exist to adjust the probabilities used in the calculations in order to correct for this systematic underestimation.

Can someone guarantee me that my insurance company uses a corrected statistic instead of the original one? I’m afraid they think I’ll die 10 years earlier than statistically expected and thus charge me too much money?!

Thursday, February 18, 2010

Life Expectancy: The difference between male and female

In this post on FlowingData, some interesting statistics are shown about life expectancy in the US.

Ever wondered why insurance companies have higher rates for men than women? The main reason is that on average, women live longer than men. In other words, the risk of dying for a man of, say 50, is higher than for a woman of that age.

Why is this? I have some ideas, but no means to prove them. What can be studied from the data is the evolution of life expectancy and the consequences of this evolution. When looking at US data for the last 20 years, simple (linear) extrapolation tells us that in 2047, men and women will have the same average age. Looking at the data for Belgium, the year is 2074.

I will probably be long dead by then, but according to this linear extrapolation, my grandchildren will have children that have life expectancies of 93 years!

I agree, linear extrapolation is an approximation. But the tendency is there: men's life expectancy is getter higher at a slightly faster pace than women's.

More about this topic (including some cool graphs) later.

Custom Search