# How long will I live?

## 17 May 2009

Input your birthday into Wolfram Alpha along with search-term `life expectancy` and it will give you an answer rounded to two decimal places. My life expectancy is 81.58 years. You can take that number and add it to your date of birth, and Wolfram Alpha will return the appropriate date. According to the databases, I can expect to live at least until 6:43:12am on Wednesday August 30 2056, which is 47 years, 3 months, 12 days from now.

Forty-seven years left, meaning the days ahead are still greater than those behind.

Interestingly, the results skew with dates going further back in time. My grandmother, for example, is 96. Born in 1913, if I do the search query with her birthdate, I get a life expectancy of 99.31 years. She has an 83.3% chance of living past 97. Adding 99.31 years to her birthdate gives us an exact date of Sunday 29 April 2012 11:02:24am.

However, the life expectancy stats in 1913 did not project at 99.31 year lifespan. I searched for this info on Alpha, and got a ‘no data available’ message. A simple search `what was the life expectancy in 1913?` gave me an answer for Canadian life expectancy in 1921 (presumably this is the earliest year for which the data is available). That result is 57.02 years.

Presuming then you were born on July 1st 1921. What is July 1 1921+57.02 years? Saturday July 8 1978. What does `life expectancy July 1 1921` give us? 93.32 years.

 birthdate: 1921-07-01 expected in that year: 57.02 expected today: 93.32

Thus the statistical life expectancy in 1921 of someone born in that year was 57 years. Alpha is clearly interpreting these queries as coming from living persons, and thus is saying, ‘if you birthdate was July 1st 1921’ then you can expect to live to be 93. You have a 76.1% chance of living past 90 and a 69.8% chance of dying before age 95.

Thus the data in 1921 has skewed forward by 36.3 years, representative of the 20th Century’s extension of the lifespan.

Another sample of search-terms:

``````life expectancy july 1 1930
life expectancy july 1 1940
life expectancy july 1 1950
etc.``` ```

birthdate age expct (current) age expct (in year-of-birth) diference
1900-07-01110.4data not availablen/a
1910-07-01 101.4 data not available n/a
1920-07-01 93.94 data not available n/a
1930-07-01 88.81 58.96 29.86
1940-07-01 85.54 64.01 21.53
1950-07-01 83.54 68.28 15.26
1960-07-01 82.38 71.04 11.34
1970-07-01 81.76 72.65 9.11
1980-07-01 81.4 75.14 6.26
1990-07-01 81.06 77.51 3.55
2000-07-01 80.89 79.42 1.47
2010-07-01 80.80 80.36 -0.44
2020-07-01 80.90 80.36 -0.54

What this shows us is that those born in the 1930 and the 1940 currently have skewed data: in the case of the 1930s, they’ve already lived thirty years more than expected when they were born, and those of the 1940s by 20 years. My own cohort (1970s) has already increased by 9 years.

If we apply the difference already for those born in the 1930s to those born in the 1970s, (and specifically for myself): 73.49+29:86 = 103.35.

In that case, I can expect to live to exactly 6:00:00pm, Tuesday June 7 2078, which is 69 years from now.

### Update September 2015

My grandmother mentioned above passed away at age 102 on 2 June 2015, outliving the 2009 projected 99.31 (29 Apr 2012) by three years.

Document History
1. Published to my blog, 17 May 2009
2. Sept 2015: this version produced