Spline extension
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Spline extension

Dear friends,
The the latest version of ASREML contains an extension to the spline
option which you should be aware of.  I am indebted to Robin Thompson
and Ian White for working out the details.

The syntax   

Y ~ mu X !r spl(X)   

will now produce a spline with no more than 50
knot points (previously the default was 100).  
If there are 50 or more unique X values, ASREML will use 50 equi-distant
knot points for the spline ranging from min(X) to max(X) instead of
using the actual data values as knot points.

The value of 50 can be changed by use of the !KNOTS k  qualifier.

datafile   !knots 15
Y ~ mu X  !r  spl(X)

would form a spline on actual unique X values if there are less < 15 such 
  otherwise the spline would be on 15 eqidistant knot points.
The former  !SPLINE K   qualifer has been redefined to
   !SPLINE  spl(X,k)  <k explicit knot points>
   where this !SPLINE statement MUST appear on the line immediately
   after the data filename line and before the linear model line.
   k is the explixit number of knot points and
   <k explicit knot points>  is the explicit set of knot points which the user
           MUST supply.  E.g.
datafile   !knots 25
 !SPLINE spl(X,9) 1 3 5 7 10 13 15 17 19
Y ~ mu X TREAT X.TREAT !r  spl(X) TREAT.spl(X,9)

would form an overal spline on actual unique X values if there are less < 25
 such values;  otherwise the spline would be on 25 eqidistant knot points;
 and an interaction spline component based on the 9 explicitly nominated 
 knot points.  When supplying knot points, you must ensure they provide
  reasonable coverage of the actual values in the variate.
  If you want to force the spline to have a particular number of equidistant
  knot points ranging min(X) to max(X), use the spl(X,k) notation without
  a preceeding !SPLINE definition.
datafile   !knots 25
Y ~ mu X TREAT X.TREAT !r  spl(X,8) TREAT.spl(X,8)

forms both spline components on 8 equidistant knot points 
[!knots 25  is ignored in this example]
I trust you are all well.


Arthur Gilmour PhD                  mailto:Arthur.Gilmour@agric.nsw.gov.au
Principal Research Scientist (Biometrics)            fax: <61> 2 6391 3899
NSW Agriculture                                           <61> 2 6391 3922
Orange Agricultural Institute             telephone work: <61> 2 6391 3815
Forest Rd, ORANGE, 2800, AUSTRALIA                  home: <61> 2 6362 0046

ASREML is still free by anonymous ftp from pub/aar on ftp.res.bbsrc.ac.uk
    or point your web browser at ftp://ftp.res.bbsrc.ac.uk/pub/aar/ 

To join the asreml discussion list, send the message  

To send messages to the list, mailto:asreml@chiswick.anprod.CSIRO.au

Asreml list archive: http://www.chiswick.anprod.csiro.au/lists/asreml

                        <> <> <> <> <> <> <>
     While they watched, He was taken up, 
          and a cloud received Him out of their sight.
     And while they looked steadfastly toward heaven as He went up, 
          behold two men stood by them in white apparel who also said
     "Men of Galilee, why do you stand gazing up into heaven?
          This same Jesus who was taken up from you into heaven,
     will so come in like manner as you saw Him go into heaven."  
                                                              Acts 1 9:11

Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml