Summary:
The SegReg computer program (model) is designed to perform a segmented linear
regression of one dependent variable (Y, e.g. plant growth, crop yield) on
one (X) or two (X and Z) independent variables (e.g. crop
growth factors like depth of water table and soil salinity).
|
Details:
The segmentation is done by introducing a breakpoint
(break-point, threshold). Thus one can obtain a broken, discontinuous, line.
Seven types of functions (0 to 6) are used. Examples are given.
The selection of the best function type and breakpoint is based on
maximizing the statistical coefficient of explanation (determination) and
performing the test of significance.
The 90% confidence interval (belt) is given as well as an Anova table, the
latter since 9 October 2005 thanks to the suggestion of Mr. Paul Barrett
(www.pbarrett.net), Psychometrician, University of Auckland, New Zealand.
On 3 March 2006 the type 2 function was improved thanks to the comments of
Mr. Francisco Leotte, National Fisheries Research Institute, Lisbon, Portugal.
In October 2007 Mr. Sergio Alonso Oroza drew my attention to a dot/comma decimal
problem and the awkward scaling in certain graphs. This was corrected.
More details are given in the program itself.
|
Start:
The mathematical model starts clicking on SegRegW.Exe.
|
Documentation:
A lecture note ("Data Analysis") on statistical analysis with examples of SegReg
applications is found on the articles page.
A paper on the statistical principles of segmented regression with break-point
can be inspected here.
The principles of regression analysis in general can be found in a lecture note
on the FAQ's page.
Also, analysis of variance (Anova) and the F-test for
segmented linear regression with break-point is briefly discussed in a paper
on the FAQ's page.
|
|
Download:
SegRegW
Go to:
software
& models
articles
& manuals
reports
case studies
FAQ's
& papers
home
page
|
