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? [[alternative HTML version deleted]]
Duncan Murdoch
2006-Jan-17 14:01 UTC
[R] how can i locate the source code of a module quickly?
On 1/17/2006 8:33 AM, obeeker at gmail.com 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