Dru Posted June 17, 2004 Posted June 17, 2004 snug fitting = it aint gonna be july 1 - that is a holiday and i will be out sending... also there are a couple more of those mysterious gaps coming up.... Quote
Gary_Yngve Posted June 20, 2004 Posted June 20, 2004 One could fit least squares polynomials in Matlab by solving A^T*Ax = A^T*b where A is an m x n matrix of m explanatatory variables of degree n , A^T is the transpose of A, b is an n-vector of y values, and x is an m-vector of coefficients (the least squares coefficients). A real matlab user would just say A\b. Quote
Gary_Yngve Posted June 20, 2004 Posted June 20, 2004 Using a 6th-degree polynomial to model that curve is lame. The fit is snug because it's overfit, and there's absolutely no justification for why the high-degree poly model is appropriate. Prove it. Take the first half of data. Build a 6th-degree poly fitting it. Extrapolate the data and see how the predictions compare to the second half of the data. Notice how the predictions suck. QED Quote
EWolfe Posted June 21, 2004 Posted June 21, 2004 snug fitting = it aint gonna be july 1 - that is a holiday and i will be out sending... also there are a couple more of those mysterious gaps coming up.... The ones that coincide with full moons? Squamish Campers beware! Quote
ashw_justin Posted June 21, 2004 Posted June 21, 2004 Using a 6th-degree polynomial to model that curve is lame. The fit is snug because it's overfit, and there's absolutely no justification for why the high-degree poly model is appropriate. Prove it. Take the first half of data. Build a 6th-degree poly fitting it. Extrapolate the data and see how the predictions compare to the second half of the data. Notice how the predictions suck. QED ok fine, you get 10 points for your post. But you lose 9 of them for ignoring that the 6th-deg poly fit was just a joke!! Well you get them back if you do a periodic (sin/cos function) fit on {Dru} because it will probably be the most accurate prediction!! Quote
Dru Posted July 6, 2004 Posted July 6, 2004 looks like its gonna be tomorrow...place bets as to time... post bets here so i can collude with the eventual winner timmy are you gonna hack the post count again? back to 9999 or back to 1? Quote
Off_White Posted July 6, 2004 Posted July 6, 2004 timmy are you gonna hack the post count again? back to 9999 or back to 1? I think it should be forward to 1,000,000. Quote
badvoodoo Posted July 7, 2004 Posted July 7, 2004 I'd have to agree, it should be ONE MILLION POSTS. Quote
icegirl Posted July 7, 2004 Posted July 7, 2004 today today today.... in one post! That's my guess between 12:30 and 1pm what do I win??!?!?! Quote
Dru Posted July 7, 2004 Posted July 7, 2004 let me tell you all this once you hit 20, 000 posts it's like finding out there is no spoon Quote
sk Posted July 8, 2004 Posted July 8, 2004 let me tell you all this once you hit 20, 000 posts it's like finding out there is no spoon sorry I missed your big day dru Quote
TrogdortheBurninator Posted August 30, 2005 Posted August 30, 2005 Yeah and it's too bad that excel can't do sine/cosine wave fits. Amatuers... Excel can do almost any fit you can imagine (linear and non linear) if you just make use of the linear regression add on and define you own variables for the regression. Quote
snugtop Posted August 30, 2005 Posted August 30, 2005 What happened to jjd? jjd where are thee? Quote
Dru Posted August 30, 2005 Posted August 30, 2005 It's gonna be approximately 6 months from now according to my highly scientific calculations. Maybe a year if my new responsibilities reduce my posting time. Quote
olyclimber Posted August 30, 2005 Posted August 30, 2005 It's gonna be approximately 6 months from now according to my highly scientific calculations. Maybe a year if my new responsibilities reduce my posting time. You have to be responsible ??!!! neener-neener-neener!!! poor sucker. Quote
sobo Posted August 30, 2005 Posted August 30, 2005 My recent calculations here indicate that Dru will top 30k in approximately 23 weeks, or just under his own estimate of 6 months. Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.