Skip to main content
gamedev.net
🔒 Locked

standard deviation

Started by davidko 6 replies 1.6k views
Original Post
davidko
davidko
My statistics really sux, so here''s my question. To calculate standard deviation for a set of points, you take the mean, find the variance, and square root it, etc. etc. But...how would you find the standard deviation for each individual point? I believe it is a measure of uncertainty for each point, but I didn''t think you could do that. An algorithm (linear regression) calls for these values, but I wasn''t sure how to calculate it...oh how I wish I had a statistics book somewhere lying about. Thanks for any help...
davidko
davidko
ahhh...

ZScore. I have absolutely no recollection of that word from stats. musta been sleeping or something. thanks!
DDnewbie
DDnewbie
Hi,

I don't quite understand what is meant by standard deviation of ONE point!??

Isn't standard deviation a measure of how a collection of points is spread around the mean? That is, a "spreadingmeasure".

You can of course calc the standard deviation of one stochastic variable, but that's a function and not a point.




/Mankind gave birth to God.

Edited by - silvren on October 16, 2001 4:48:28 PM
/Mankind gave birth to God.
Zipster
Zipster
The same thing happens when working with derivatives... how can you find the slope of the line tangeant to the function at a point when we need two points to define a line? We can of course do this, but it will always be off by an infinitely small amount. I don''t recall anything about ZScore either, so I can''t directly tell you how it applies in that sense, but it''s somewhat related!
Beer Hunter
Beer Hunter
Zipster - yes, it will be off by an infinitely small amount. Keyphrase: "infinitely small".
DDnewbie
DDnewbie
OK,

Thanks! the concept is somewhat clearer to me now.
I think I''ll read up on that stuff.

cheers

/Mankind gave birth to God.
/Mankind gave birth to God.

Topic Locked

This topic has been locked by a moderator. New replies are not allowed.

Sign in to reply to this topic.