Re: Negative R2

OK. Thank you.

Well I have not skipped books but I have just referred to applied
statistics book which did not discuss nitty-gritty of all the methods
and my main field is not statistics. I also checked this group for
past discussions on negative R2 and there are many opinions. I would
appreciate your suggestions on good statistics books.

Now let me give you background about the question I am checking my
model on an external dataset in the sense I have not used this
external dataset to make my model. So the objective is to check the
applicability of the model capable of reproducing the experimental
data and not just producing predicted activity values that correlate
with the experimental data.

Deep

On Feb 4, 4:33 pm, "Old Mac User" <chendrixst...@xxxxxxxxx> wrote:
Fitting a line through the origin is a bad practice for several
reasons. One reason is that yes, indeed, you can end up with anegativeR-sq. The other reasons are more subtle, and much more
dangerous. Just don't fit a line running through the origin, also
called a "no intercept model".

It you are just starting to learn statistics, then it seems to me that
you have skipped over most of the basics and are rushing headlong into
using methods that you do not understand. I suggest you go back to the
beginning and start all over without using software that you don't
understand. OMU

On Feb 3, 6:27 pm, "Deep" <Deepangi.Pan...@xxxxxxxxx> wrote:



Hello Group:
I am learning statistics and new to its subtleties. I am using partial
least square (PLS) method to develop my model. It is suggested in
literature to check the stability of model by plotting Predicted vs.
Actual and Actual vs. Predicted and for both also check fit line
through the origin. However I getnegativeR2(rsquare) for line
fitted through origin for plot y= actual and x= predicted. Could you
please explain the reason? I understand that this line does not fit in
a linear fashion and may be it is curved but I would appreciate
insight from the group members.

Thank you.
Deep- Hide quoted text -

- Show quoted text -


.

Relevant Pages

  • Re: [OT][Long][You All Know Everything Dept] Recommendations for Stat book?
    ... > small in size, but pithy in ... > explanation. ... I have a statistics book which, based on random sampling of statistics ...
    (comp.lang.lisp)
  • Re: Photons (something for Porat)
    ... There are two reasons to invent photons; ... that there is a relation and quantisation could ... The use of statistics is a good way to get 'grip' on a continuum. ... As this has a timelike axis, we could turn that axis into a light-like direction and turn the circle into a helix. ...
    (sci.physics.relativity)
  • Re: I cannot understand the sentence in a probability book.
    ... But finitely additivity ... are not useful in statistical problems for many reasons. ... Herman Rubin, Department of Statistics, Purdue University ...
    (sci.stat.math)
  • Re: Photons (something for Porat)
    ... There are two reasons to invent photons; ... Actually h has the same dimension as angular momentum and that relates to a cross product, like the curl in Maxwells equations. ... The use of statistics is a good way to get 'grip' on a continuum. ...
    (sci.physics.relativity)
  • Re: Aether is the empty space in which the Universe sits
    ... in my genealogy I have many direct ancestors who lived ... Go by the statistics. ... Life expetancy has doubled since 1900. ... reasons is hygene (which the ancients did not practice: ...
    (sci.physics.relativity)