Re: Happy Pi Day, 2021

Posted by Jon Peck on
URL: http://spssx-discussion.165.s1.nabble.com/Happy-Pi-Day-2021-tp5740284p5740291.html

What the article doesn't mention - and this is a more practical consideration - is the accuracy of the modeling data that pi might be applied to.  Most measurements in practice are probably not good to more than four or five significant figures, if that, at least in the social sciences.  I used to give students as an exercise to add a random digit at the accuracy limit and see how it affected their model estimates.  Throw in a little multicollinearity and there you are.

See the famous Morgenstern book  A Perspective on the Accuracy of Economic Observations on JSTOR  and some updates (behind a paywall, unfortunately)

On Mon, Mar 15, 2021 at 9:38 AM Andy W <[hidden email]> wrote:

Well Bruce, at least your solution is good enough for government work,
https://www.jpl.nasa.gov/edu/news/2016/3/16/how-many-decimals-of-pi-do-we-really-need/.



-----
Andy W
[hidden email]
http://andrewpwheeler.wordpress.com/
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

--
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD