On 1/17/2006 8:33 AM,

[hidden email] wrote:

> I have dowloaded the Source Code of R,and I want to know the process of

> chi-sqared test,but how can I found it?

Which function are you using? Let's assume chisq.test. Then the first

thing to do is to type the function name:

> chisq.test

function (x, y = NULL, correct = TRUE, p = rep(1/length(x), length(x)),

rescale.p = FALSE, simulate.p.value = FALSE, B = 2000)

{

DNAME <- deparse(substitute(x))

[ lots of skipped lines ]

structure(list(statistic = STATISTIC, parameter = PARAMETER,

p.value = PVAL, method = METHOD, data.name = DNAME, observed = x,

expected = E, residuals = (x - E)/sqrt(E)), class = "htest")

}

<environment: namespace:stats>

>

This is a deparsed version of the source; often it's good enough. If

not, it tells you that the function is in the stats package, so you can

look for the original source (with comments, if you're lucky) in

src/library/stats/R.

For some functions you get less information:

> pchisq

function (q, df, ncp = 0, lower.tail = TRUE, log.p = FALSE)

{

if (missing(ncp))

.Internal(pchisq(q, df, lower.tail, log.p))

else .Internal(pnchisq(q, df, ncp, lower.tail, log.p))

}

<environment: namespace:stats>

The .Internal() call tells you that this function is mostly implemented

in C or Fortran code from the R source. (.C() or .Fortran() or .Call()

would tell you that the source is in the package src/library/stats/src

directory).

To find its source, look in src/main/names.c. There's a huge table

there, containing these lines:

{"dchisq", do_math2, 6, 11, 2+1, {PP_FUNCALL, PREC_FN, 0}},

{"pchisq", do_math2, 7, 11, 2+2, {PP_FUNCALL, PREC_FN, 0}},

{"qchisq", do_math2, 8, 11, 2+2, {PP_FUNCALL, PREC_FN, 0}},

These tell you that the chisq functions are all implemented in a

function called do_math2, with codes 6, 7, and 8 respectively. (Further

up in names.c is a comment describing the other columns.)

You need to do a grep or other search to find do_math2; it's in

src/main/arithmetic.c. You can follow the source from there.

Sometimes the source to the function won't tell you which package it is

in (because the package doesn't have a namespace defined). In that case

the "getAnywhere" function can be useful, e.g.

> getAnywhere("airmiles")

A single object matching 'airmiles' was found

It was found in the following places

package:datasets

with value

Time Series:

Start = 1937

End = 1960

Frequency = 1

[1] 412 480 683 1052 1385 1418 1634 2178 3362 5948 6109

5981

[13] 6753 8003 10566 12528 14760 16769 19819 22362 25340 25343 29269 30514

airmiles isn't a function, it's a dataset, and it is located in

src/library/datasets/data/airmiles.R.

I hope this helps.

Duncan Murdoch

______________________________________________

[hidden email] mailing list

https://stat.ethz.ch/mailman/listinfo/r-helpPLEASE do read the posting guide!

http://www.R-project.org/posting-guide.html