search for: fmax2

Dear R list members, I have a question about the proper use of the typsize and fscale arguments to nlm. I use nlm in my sem package to fit general structural-equation models, which entails maximizing a multinormal likelihood with respect to parameters that represent regression coefficients and covariances of variables. The magnitudes of these parameters can be very different. The

...nd find the largest abs(entry) */ + /* first check row for any NAs and find the largest entry */ large = 0.0; isna = FALSE; for (c = 0; c < *nc; c++) { a = matrix[r + c * n_r]; if (ISNAN(a)) { isna = TRUE; break; } - if (do_rand) large = fmax2(large, fabs(a)); + if (!R_FINITE(a)) continue; + if (do_rand) large = fmax2(large, a); } if (isna) { maxes[r] = NA_INTEGER; continue; } ---------------- END ---------------------- This gives the expected behavior in the two examples above. (Sorry to crosspost...

Full_Name: Wolfgang Resch Version: R 2.8.1 GUI 1.27 OS: OS X 10.4.11 Submission from: (NULL) (137.187.89.14) Strange behavior of qbinom: > qbinom(0.01, 5016279, 1e-07) [1] 0 > qbinom(0.01, 5016279, 2e-07) [1] 16 > qbinom(0.01, 5016279, 3e-07) [1] 16 > qbinom(0.01, 5016279, 4e-07) [1] 16 > qbinom(0.01, 5016279, 5e-07) [1] 0

...kup -single_module -multiply_defined suppress -L/usr/local/ lib -o pnmath.so bd0.o beta.o chebyshev.o choose.o d1mach.o dbeta.o dbinom.o dcauchy.o dchisq.o dexp.o df.o dgamma.o dgeom.o dhyper.o dlnorm.o dlogis.o dnbeta.o dnbinom.o dnchisq.o dnf.o dnorm.o dnt.o dpois.o dt.o dunif.o dweibull.o fmax2.o fmin2.o ftrunc.o gamma.o gamma_cody.o i1mach.o imax2.o imin2.o lbeta.o lgamma.o lgammacor.o mlutils.o pbeta.o pbinom.o pcauchy.o pchisq.o pexp.o pf.o pgamma.o pgeom.o phyper.o plnorm.o plogis.o pnbeta.o pnbinom.o pnchisq.o pnf.o pnmath.o pnorm.o pnt.o polygamma.o ppois.o pt.o ptukey.o pun...

...binom.c @@ -104,6 +104,7 @@ double qnbinom(double p, double size, double prob, int lower_tail, int log_p) /* y := approx.value (Cornish-Fisher expansion) : */ z = qnorm(p, 0., 1., /*lower_tail*/TRUE, /*log_p*/FALSE); y = R_forceint(mu + sigma * (z + gamma * (z*z - 1) / 6)); + y = fmax2(0.0, y); z = pnbinom(y, size, prob, /*lower_tail*/TRUE, /*log_p*/FALSE); I used the https://github.com/r-devel/r-svn repo and its continuous integration tools to check that it doesn't break any existing tests: https://github.com/r-devel/r-svn/actions/runs/201327042 I have also requested...

...src/nmath/qnbinom.c b/src/nmath/qnbinom.c index b313ce56b2..16845d9373 100644 --- a/src/nmath/qnbinom.c +++ b/src/nmath/qnbinom.c @@ -49,10 +49,18 @@ do_search(double y, double *z, double p, double n, double pr, double incr) { if(*z >= p) { /* search to the left */ for(;;) { + y = fmax2(0, y - incr); if(y == 0 || - (*z = pnbinom(y - incr, n, pr, /*l._t.*/TRUE, /*log_p*/FALSE)) < p) - return y; - y = fmax2(0, y - incr); + (*z = pnbinom(y, n, pr, /*l._t.*/TRUE, /*log_p*/FALSE)) < p){ + if(incr == 1){ + // we know that the sear...

...nd the largest abs(entry) */ + /* first check row for any NAs and find the largest entry */ large = 0.0; isna = FALSE; for (c = 0; c < *nc; c++) { a = matrix[r + c * n_r]; if (ISNAN(a)) { isna = TRUE; break; } - if (do_rand) large = fmax2(large, fabs(a)); + if (!R_FINITE(a)) continue; + if (do_rand) large = fmax2(large, a); } if (isna) { maxes[r] = NA_INTEGER; continue; } --------------------------------------- END ------------------------------------------------- This gives the expected behavio...

...e300, n = 1e9, min.n = 1) gave 20 intervals, far from 1e9, but pretty(c(-1,1)*1e300, n = 1e6, min.n = 1) gave 1000000 intervals (on a machine), made me trace through the code to function 'R_pretty' in https://svn.r-project.org/R/trunk/src/appl/pretty.c . *lo is -1e300, *up is 1e300. cell = fmax2(fabs(*lo),fabs(*up)); 'cell' is 1e300. i_small = dx < cell * U * imax2(1,*ndiv) * DBL_EPSILON *3; When *ndiv is (int) 1e9, apparently cell * U * imax2(1,*ndiv) overflows to infinity and 'i_small' is 1 (true). It doesn't happen when *ndiv is (int) 1e6. Putting parentheses may...

Bug Description: Problem with the function pf() when the non-centrality parameter is large. Here is a sample command. You should see a smooth line from 0 to about 55, and then the values of pf() go crazy from 55 to 100. ############################ ncp <- seq(0,100,length=200) plot(ncp,pf(5,7,2,ncp=ncp)) ############################ Version: platform = i686-pc-linux-gnu arch = i686 os =

...t; pretty(c(-1,1)*1e300, n = 1e6, min.n = 1) gave 1000000 intervals ? ? >> (on a machine), made me trace through the code to function 'R_pretty' in https://svn.r-project.org/R/trunk/src/appl/pretty.c . ? ? > thank you. ? ? >> *lo is -1e300, *up is 1e300. ? ? >> cell = fmax2(fabs(*lo),fabs(*up)); ? ? >> 'cell' is 1e300. ? ? >> i_small = dx < cell * U * imax2(1,*ndiv) * DBL_EPSILON *3; ? ? >> When *ndiv is (int) 1e9, apparently cell * U * imax2(1,*ndiv) overflows to infinity and 'i_small' is 1 (true). It doesn't happen when *ndiv...

...src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -g -O2 -c wilcox.c -o wilcox.o gcc -std=gnu99 -I. -I../../src/include -I../../src/include -I/usr/local/include -DHAVE_CONFIG_H -g -O2 -c signrank.c -o signrank.o rm -rf libnmath.a ar cr libnmath.a mlutils.o d1mach.o i1mach.o fmax2.o fmin2.o fprec.o fround.o ftrunc.o sign.o fsign.o imax2.o imin2.o chebyshev.o log1p.o expm1.o lgammacor.o gammalims.o stirlerr.o bd0.o gamma.o lgamma.o gamma_cody.o beta.o lbeta.o polygamma.o bessel_i.o bessel_j.o bessel_k.o bessel_y.o choose.o snorm.o sexp.o dgamma.o pgamma.o qgamma.o rgamma.o...

...return A*(1.13113163544180*u1+u2-1); + /** <modified constant> **/ + return A*(1.131131635444180*u1+u2-1); + /** </modified> **/ } if(u1 >= 0.973310954173898) { /* tail: */ @@ -241,6 +243,10 @@ tt = 0.479727404222441-0.595507138015940*fmin2(u2,u3); if(fmax2(u2,u3) <= 0.805577924423817) return (u2<u3) ? tt : -tt; + /** <added> **/ + if(0.053377549506886*fabs(u2-u3) <= g(tt)) + return (u2<u3) ? tt : -tt; + /** </added> **/ } case BOX_MULLER: if(BM_norm_keep != 0.0) { /* An exact test is intentional *...

...uble qnbinom(double p, double size, double prob, > int lower_tail, int log_p) > /* y := approx.value (Cornish-Fisher expansion) : */ > z = qnorm(p, 0., 1., /*lower_tail*/TRUE, /*log_p*/FALSE); > y = R_forceint(mu + sigma * (z + gamma * (z*z - 1) / 6)); > + y = fmax2(0.0, y); > z = pnbinom(y, size, prob, /*lower_tail*/TRUE, /*log_p*/FALSE); > I used the https://github.com/r-devel/r-svn repo and its continuous > integration tools to check that it doesn't break any existing tests: > https://github.com/r-devel/r-svn/actions/runs/2...

Full_Name: Morten Welinder Version: 1.6.1 OS: Solaris/sparc Submission from: (NULL) (65.213.85.144) qbeta(0.1, 1e-8, 0.5, TRUE, FALSE) seems to hang for me.

...from 1e9, but > pretty(c(-1,1)*1e300, n = 1e6, min.n = 1) gave 1000000 intervals > (on a machine), made me trace through the code to function 'R_pretty' in https://svn.r-project.org/R/trunk/src/appl/pretty.c . thank you. > *lo is -1e300, *up is 1e300. > cell = fmax2(fabs(*lo),fabs(*up)); > 'cell' is 1e300. > i_small = dx < cell * U * imax2(1,*ndiv) * DBL_EPSILON *3; > When *ndiv is (int) 1e9, apparently cell * U * imax2(1,*ndiv) overflows to infinity and 'i_small' is 1 (true). It doesn't happen when *ndiv is (int) 1e...

...o tabulate.o uncmin.o zeroin.o ch2in v.o chol.o dpbfa.o dpbsl.o dpoco.o dpodi.o dpofa.o dposl.o dqrdc.o dqrdc2.o dqrls.o dqrsl.o dqrutl.o dsvdc.o dtrco.o dtrsl.o eigen.o lminfl.o : libappl.a making mlutils.d from mlutils.c making d1mach.d from d1mach.c making i1mach.d from i1mach.c making fmax2.d from fmax2.c making fmin2.d from fmin2.c making fprec.d from fprec.c making fround.d from fround.c making ftrunc.d from ftrunc.c making sign.d from sign.c making fsign.d from fsign.c making imax2.d from imax2.c making imin2.d from imin2.c making chebyshev.d from chebyshev.c making log1p.d from lo...

...mfa5/gilp/R-versions/R-devel/src/appl' make[2]: Entering directory `/home/mfa5/gilp/R-versions/R-devel/src/nmath' make[3]: Entering directory `/home/mfa5/gilp/R-versions/R-devel/src/nmath' making mlutils.d from mlutils.c making d1mach.d from d1mach.c making i1mach.d from i1mach.c making fmax2.d from fmax2.c making fmin2.d from fmin2.c making fmod.d from fmod.c making fprec.d from fprec.c making fround.d from fround.c making ftrunc.d from ftrunc.c making sign.d from sign.c making fsign.d from fsign.c making fsquare.d from fsquare.c making fcube.d from fcube.c making imax2.d from imax2.c...

...int current_pos) { double a, p, y, num, denom, leftval, rtval; a = node[current_pos + 1]; y = node[current_pos] + node[current_pos+2]; num = accept_prob_fn(a, y); leftval = accept_prob_fn(a, -theta + node[current_pos+2]); rtval = accept_prob_fn(a, theta + node[current_pos+2]); denom = fmax2(leftval, rtval); /* function in Rmath which returns max of two doubles */ p = num/denom; assert(p>=0 && p <=1); return p; } /* auxilary function for accept_prob */ double accept_prob_fn(double a, double y) { double f; if(y != 0) f = (exp(y) - exp(-y))/(y*exp(a*y))...

...ct -g -I../include -I../../src/include -c mlutils.c -o mlutils.o gcc -ieee_with_inexact -g -I../include -I../../src/include -c d1mach.c -o d1mach.o gcc -ieee_with_inexact -g -I../include -I../../src/include -c i1mach.c -o i1mach.o gcc -ieee_with_inexact -g -I../include -I../../src/include -c fmax2.c -o fmax2.o gcc -ieee_with_inexact -g -I../include -I../../src/include -c fmin2.c -o fmin2.o gcc -ieee_with_inexact -g -I../include -I../../src/include -c fmod.c -o fmod.o gcc -ieee_with_inexact -g -I../include -I../../src/include -c fprec.c -o fprec.o gcc -ieee_with_inexact -g -I../include...

...../lib/libgraphics.a graphics.o ranlib ../lib/libgraphics.a cd nmath; make cc -ieee_with_inexact -g -I../../src/include -c mlutils.c cc -ieee_with_inexact -g -I../../src/include -c d1mach.c cc -ieee_with_inexact -g -I../../src/include -c i1mach.c cc -ieee_with_inexact -g -I../../src/include -c fmax2.c cc -ieee_with_inexact -g -I../../src/include -c fmin2.c cc -ieee_with_inexact -g -I../../src/include -c fmod.c cc -ieee_with_inexact -g -I../../src/include -c fprec.c cc -ieee_with_inexact -g -I../../src/include -c fround.c cc -ieee_with_inexact -g -I../../src/include -c ftrunc.c cc -ieee_wi...