RobertNiles.com
About Robert
Reporting Help
Finding Data on the Internet
Questions for Candidates
Stats Lessons
Mean
Median
Percent
Per capita
Standard Deviation
Margin of Error
Data Analysis
Sample Sizes
Stats Tests
Help Board
Bookstore


The Cartoon Guide to Statistics

This isn't some dumb-downed whitewash. It's a must-read for a beginner taking his or her first steps toward understanding stats.
More information
-->

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

here's a challenging one
Message posted by Neil (via 194.82.103.75) on February 28, 2002 at 9:47 AM (ET)

How do I solve the cofficients in the equation of type
y=a+b+cX+dX. by least squares where X has the same value. Ta. NH.


READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: here's a challenging one
Message posted by Darius (via 200.23.217.10) on February 28, 2002 at 10:58 AM (ET)

y=a+b+cX+dX. by least squares where X has the same value. ?????????

I tink you made a mistake, the equation:

y=a+b+cX+dX
being a+b=A and c+d = B >>> y= A+BX

A traditional univariate least squares


Re: here's a challenging one
Message posted by Jack Tomsky (via 12.144.103.66) on February 28, 2002 at 1:46 PM (ET)

If X has the same value, then the design matrix has a rank of one. What this means is that you cannot solve for a, b, c, and d separately. The least squares estimate for the linear function, a+b+cX+dX, is obtained by setting it equal to ybar.


Re: here's a challenging one
Message posted by Neil (via 194.82.103.76) on March 1, 2002 at 6:50 AM (ET)

Thanks to both of you, this problem is not mentioned in any book I can find. I did not understand Jack's response entirely but I assume it is the same as Darius's.



Your $5 contribution helps cover part the $500 annual cost of keeping this site online.

Niles Online Sites:RobertNiles.comTheme Park InsiderViolinist.com

RobertNiles.com™, the site, content and services 咀opyright 1996-2002, Robert Niles.
All rights reserved. Questions? Comments? Read my Privacy Policy, or E-mail me!