1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
|
This is an R modification by Thomas Lumley
of S code by Mike Meyer
that calls an adaptive quadrature routine by Alan Genz.
Original http://lib.stat.cmu.edu/S/adapt [dated Oct 17 1989]
-------
June 2001
Removed one-dimensional integration: R 1.3.0 has a quadpack-based function
in base.
-------
June 1999
Deleted Makefile -- R now does better without one.
Redid function passing to use eval() rather than call_R.
Added option to keep doubling static storage allocation until precision
achieved or R crashes :)
Thomas Lumley
----
January 1999:
The Fortran code works without problems
on Solaris 2.5.1 [f77 SC4.2; g77] and Linux (Debian 2.0).
There's no reason to keep the f2c-translated C code.
adapt() and hence integrate() now return a class "integrate" object
having a simple print method.
Martin Maechler, ETH Zurich
----
July 1997:
I recommend using the C code (converted with f2c) as I'm not sure about
the correct linking for the FORTRAN. If you use g77 to compile the
FORTRAN you must specify the -O flag, as otherwise bsrl.f activates a
known bug in g77 causing it to generate invalid assembly language.
In any case, make sure that the test examples work.
Thomas Lumley
thomas@biostat.washington.edu
|
