Search:  

 
 
   All ForumsHot TopicsGallery






how-to block ads


 
Forums » Backbone Analysis Puts Exaflood Myth To Bed » Past statistics don't always show future trends
Search Topic:
Share Topic:
RSS topic:
toggle:
flat / full
normal / watch
Post a:
Post a:
Woo Hoo! Free capacity! »
« My View  
AuthorAll Replies


Nightshade
sic semper tyrannis
Premium
join:2002-05-26
Salem, OR

reply to TKJunkMail
Re: Past statistics don't always show future trends

said by TKJunkMail See Profile :

Those predicting drastic growth may be more prescient than the statisticians looking backward.
I have a very serious problem with that statement and here's why.

I took statistics and probability. If there's anything I learned out of that class is that statisticians look at past data to make predictions about future trends. Without past data, there is no baseline for them to show future trends.

You can't have one without the other. You got to have past data to create any statistical trends at all regardless if the trend is accurate or not, period.
--
Be careful who you vote for, you just might get it.


TKJunkMail
Enjoy the sun
Premium
join:2002-03-03
Avalon, NJ
·Sprint Mobile Broa..
·Comcast

said by Nightshade See Profile :

I took statistics and probability. If there's anything I learned out of that class is that statisticians look at past data to make predictions about future trends. Without past data, there is no baseline for them to show future trends.

You can't have one without the other. You got to have past data to create any statistical trends at all regardless if the trend is accurate or not, period.
So did I. And there is no statistical models that project when the knee in a curve may happen. And that is what you can't predict based purely on past trends.
--
My BLOG .. .. Internet News .. .. My Web Page
Ask yourself one question: 'Do I feel lucky?' Well, do ya punk?


Nightshade
sic semper tyrannis
Premium
join:2002-05-26
Salem, OR


1 edit
Oh I know. If anything creating statistics is really a combination of mathematical theory and good old fashioned educated guessing. All statistical models will eventually fail, it's just a matter of when and how the data that already exists is modeled. But you still need strong past data to make the predictions. Without that all you really are doing is doing baseless guessing.
--
Be careful who you vote for, you just might get it.
-
Forums » Backbone Analysis Puts Exaflood Myth To BedWoo Hoo! Free capacity! »
« My View  


Friday, 04-Dec 16:09:36 Terms of Use | Privacy Policy | Hosting by www.nac.net - DSL,Hosting & Co-lo | feedback | contact
over 10 years online! © 1999-2009 dslreports.com.
page compression OFF
Most commented news this week
· [163] Comcast Releasing Promised Usage Meter
· [145] Avast Antivirus Has Gone Mad
· [121] Comcast Makes NBC Universal Acquisition Official
· [104] Graduate Student Unveils Sprint's GPS Sharing With Feds
· [101] Google Invades ISP, OpenDNS Turf With Google Public DNS
· [81] Latest Consumer Reports Survey Not Kind To AT&T
· [79] FCC Ponders Moving From PSTN To IP Voice
· [74] Sprint Defuses GPS Privacy Media Bomb
· [70] Baltimore To Ban Lazy Cable Installs
· [64] Broadband Killed The Game Console
Most people now reading
· False positive in Avast! or is it real? [Security]
· DNS options, what are YOU using? [TekSavvy]
· Maximizing Rogue DPS for ToC/ToGC (3.x) [World of Warcraft]
· 3.x Feral Druid - Bear Tanking Guide [World of Warcraft]
· People would bewalking away from the table if ACTA public [TekSavvy]
· Long ethernet runs [Wireless Service Providers]
· Connecting to Google Voice Via SIP [VOIP Tech Chat]
· Extjs grid combo box. [Webmasters and Developers]
· [Rant] Disrespect of PTO [Rants, Raves, and Praise]
· Windows 7 boot manager editing questions [Microsoft Help]