My Math Forum > Math Method of Least Squares

 Math General Math Forum - For general math related discussion and news

March 20th, 2017, 06:03 PM   #1
Newbie

Joined: Nov 2016
From: England

Posts: 2
Thanks: 0

Method of Least Squares

Hi,

I've attached the question and my worked solution.

I am confident I have the correct least squares straight line of best fit to the data to 2 d.p., but I'm unsure what is being asked in the next part.

Am I to sub the xi values into the least squares straight line of best fit to acquire yi estimates?

And if so, I know there will be variation. How would you explain the accuracy of such variation?
Attached Images
 IMAG0130_1.jpg (97.4 KB, 11 views) IMAG0132.jpg (89.1 KB, 10 views)

 March 29th, 2017, 09:10 PM #2 Newbie   Joined: Nov 2013 Posts: 29 Thanks: 1 Was this for some class?
March 29th, 2017, 10:16 PM   #3
Senior Member

Joined: Jul 2008

Posts: 4,027
Thanks: 39

The parameters that you calculated for your regression line look correct.

Quote:
 Originally Posted by ChrisHughes Am I to sub the xi values into the least squares straight line of best fit to acquire yi estimates?
Yes, that's how I understand the question. Not sure what they want you to comment on exactly. Sometimes, the data will fit the regression line very closely, and sometimes it won't. I'm not sure there's much more that you can say. If there were some outliers, you could comment on that, but I don't see any.

 March 29th, 2017, 10:20 PM #4 Senior Member     Joined: Sep 2015 From: USA Posts: 1,931 Thanks: 999 you've estimated a linear model given this data. $y(x) = 8.71x + 31.91$ for each $x_i$ find $y(x_i) = 8.71x_i + 31.91$ Plot these points in one color and your actual data points in another. You'll find the data isn't particularly linear in this case.
April 2nd, 2017, 04:08 PM   #5
Senior Member

Joined: Oct 2013
From: New York, USA

Posts: 600
Thanks: 82

Quote:
 Originally Posted by romsek you've estimated a linear model given this data. $y(x) = 8.71x + 31.91$ for each $x_i$ find $y(x_i) = 8.71x_i + 31.91$ Plot these points in one color and your actual data points in another. You'll find the data isn't particularly linear in this case.
In order to provide as much help to ChrisHughes as possible, we need to know if he has studied regression that isn't linear.

 Tags method, squares

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post smslca Number Theory 3 April 7th, 2011 04:43 PM elektron1999 Applied Math 1 November 20th, 2010 01:06 PM elektron1999 Linear Algebra 0 November 20th, 2010 12:03 PM WillingSponge Number Theory 5 July 9th, 2010 06:52 AM skully_888 Applied Math 0 March 19th, 2007 11:33 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top