thr3ads.net - R help - [R] Strange means of numbers drawn from rpois [Nov 2000]

If this information is useful, please help other people find it:
Share via:

Kjetil Kjernsmo

2000-Nov-08 14:09 UTC

[R] Strange means of numbers drawn from rpois

Dear all,

I think I must be going crazy.... If I do> mean(rpois(1000000, 14))I get:
[1] 13.50420
and again:> mean(rpois(1000000, 14))
[1] 13.49896> mean(rpois(1000000, 14))
[1] 13.50161> mean(rpois(1000000, 15))
[1] 14.49250> mean(rpois(1000000, 15))
[1] 14.49897> mean(rpois(1000000, 14.5))
[1] 13.99689> mean(rpois(1000000, 14.5))[1] 13.9963

This was on a
platform alphaev6-dec-osf4.0e
major    1
minor    1.1
checking out on a
platform sparc-sun-solaris2.7
major    1
minor    1.0
I get the same> mean(rpois(100000, 14.5))
[1] 14.00347> mean(rpois(100000, 14.5))
[1] 14.00667> mean(rpois(100000, 14.5))[1] 14.00675

It's fishy, isn't it...? I have generated lots of numbers, but I
haven't
seen anything like this before....

I have also done
> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  1.000052  1.999150  2.996975  4.003843  4.996136  5.999065  6.999725
 [8]  7.998869  9.003127  9.518397 10.512347 11.500405 12.510494 13.499788
[15] 14.504661

So, something happens at 10... And again:
> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  1.000173  2.001639  2.997016  4.000163  5.000802  5.992751  6.999794
 [8]  7.997315  8.998479  9.517824 10.508328 11.507336 12.501885 13.502146
[15] 14.496810

Starting with --vanilla gives pretty much the same result...

trying different RNGs:> RNGkind("Mersenne")
> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  1.000563  1.998802  2.998526  3.997537  4.999230  5.996591  6.999596
 [8]  7.998982  9.000783  9.510220 10.509562 11.505009 12.505001 13.504522
[15] 14.500956> RNGkind("Knut")
> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  0.999031  2.000444  3.002296  3.997795  4.998880  5.995267  7.001894
 [8]  7.999983  9.000267  9.513537 10.507142 11.502634 12.498944 13.498615
[15] 14.502410

After printing rpois (to see that I haven't replaced it with something
strange) and version, I call> plot(sapply(1:15, function(n) mean(rpois(1000000, n))), t='l')
getting a plot that isn't a straight line, and then:> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  0.997961  1.999509  3.000748  3.999825  5.002983  6.004424  6.998825
 [8]  7.997999  8.998341  9.766942 10.781751 11.817181 12.856154 13.900369
[15] 14.999378
and again:> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  0.998570  2.000155  2.999534  4.002987  4.997999  5.996326  6.996761
 [8]  7.996728  8.997265  9.764972 10.793913 11.822874 12.859886 13.903846
[15] 15.000311

Weird....?

Best,

Kjetil
-- 
Kjetil Kjernsmo
Graduate astronomy-student                    Problems worthy of attack
University of Oslo, Norway            Prove their worth by hitting back
E-mail: kjetikj at astro.uio.no                                - Piet Hein
Homepage <URL:http://www.astro.uio.no/~kjetikj/>
Webmaster at skepsis.no


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Alan Arnholt

2000-Nov-08 14:55 UTC

head link

[R] Strange means of numbers drawn from rpois

Kjetil,

I get "expected" results using a windows machine version 
1.1.1.

> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  1.000284  1.999766  3.001319  4.000618  5.000408  6.003184  7.000026 
7.997165  9.000461  9.514773 11.000396
[12] 12.128740 13.160210 14.177194 15.193355> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  1.000726  2.000776  3.000708  4.000043  5.001959  6.001235  7.002085 
7.998147  9.000528  9.901534 10.999140
[12] 12.124794 13.152148 14.182332 15.007930> sapply(1:15, function(n) mean(rpois(1000000, n))) [1]  1.000320  2.001331  3.002032  4.004577  4.999679  6.000778  6.997700 
8.002213  9.005890  9.764240 10.786886
[12] 11.822689 12.863624 13.904637 15.000833>  sapply(1:15, function(n) mean(rpois(100000, n))) [1]  1.00049  1.99307  2.99624  3.99277  5.00063
 [6]  6.00018  6.99672  8.01585  8.98921  9.76499
[11] 10.76767 11.78713 12.82248 13.85002 14.89385>  sapply(1:15, function(n) mean(rpois(100000, n))) [1]  0.99788  2.00799  3.00712  3.98751  4.99152
 [6]  5.99747  7.01120  7.99749  9.00031  9.73359
[11] 10.75011 12.00378 13.11141 14.14407 15.17481>  sapply(1:15, function(n) mean(rpois(100000, n))) [1]  1.00198  1.98872  3.00582  3.99331  4.99453
 [6]  6.01429  6.99885  7.98966  8.99976  9.87061
[11] 10.90451 11.99415 13.13388 14.15692 15.16398>  sapply(1:15, function(n) mean(rpois(100000, n))) [1]  0.99634  1.99840  3.00492  3.99907  4.99878
 [6]  6.00134  7.00225  7.98363  8.99210  9.85822
[11] 10.89415 12.00827 13.12437 14.16350 14.99912



On Wed, 8 Nov 2000 15:09:24 +0100 (MET) Kjetil Kjernsmo 
<kjetil.kjernsmo at astro.uio.no> wrote:
> Dear all,
> 
> I think I must be going crazy.... If I do
> > mean(rpois(1000000, 14))
> I get:
> [1] 13.50420
> and again:
> > mean(rpois(1000000, 14))
> [1] 13.49896
> > mean(rpois(1000000, 14))
> [1] 13.50161
> > mean(rpois(1000000, 15))
> [1] 14.49250
> > mean(rpois(1000000, 15))
> [1] 14.49897
> > mean(rpois(1000000, 14.5))
> [1] 13.99689
> > mean(rpois(1000000, 14.5))
> [1] 13.9963
> 
> This was on a
> platform alphaev6-dec-osf4.0e
> major    1
> minor    1.1
> checking out on a
> platform sparc-sun-solaris2.7
> major    1
> minor    1.0
> I get the same
> > mean(rpois(100000, 14.5))
> [1] 14.00347
> > mean(rpois(100000, 14.5))
> [1] 14.00667
> > mean(rpois(100000, 14.5))
> [1] 14.00675
> 
> It's fishy, isn't it...? I have generated lots of numbers, but I
haven't
> seen anything like this before....
> 
> I have also done
> 
> > sapply(1:15, function(n) mean(rpois(1000000, n)))
>  [1]  1.000052  1.999150  2.996975  4.003843  4.996136  5.999065  6.999725
>  [8]  7.998869  9.003127  9.518397 10.512347 11.500405 12.510494 13.499788
> [15] 14.504661
> 
> So, something happens at 10... And again:
> 
> > sapply(1:15, function(n) mean(rpois(1000000, n)))
>  [1]  1.000173  2.001639  2.997016  4.000163  5.000802  5.992751  6.999794
>  [8]  7.997315  8.998479  9.517824 10.508328 11.507336 12.501885 13.502146
> [15] 14.496810
> 
> Starting with --vanilla gives pretty much the same result...
> 
> trying different RNGs:
> > RNGkind("Mersenne")
> > sapply(1:15, function(n) mean(rpois(1000000, n)))
>  [1]  1.000563  1.998802  2.998526  3.997537  4.999230  5.996591  6.999596
>  [8]  7.998982  9.000783  9.510220 10.509562 11.505009 12.505001 13.504522
> [15] 14.500956
> > RNGkind("Knut")
> > sapply(1:15, function(n) mean(rpois(1000000, n)))
>  [1]  0.999031  2.000444  3.002296  3.997795  4.998880  5.995267  7.001894
>  [8]  7.999983  9.000267  9.513537 10.507142 11.502634 12.498944 13.498615
> [15] 14.502410
> 
> After printing rpois (to see that I haven't replaced it with something
> strange) and version, I call
> > plot(sapply(1:15, function(n) mean(rpois(1000000, n))), t='l')
> getting a plot that isn't a straight line, and then:
> > sapply(1:15, function(n) mean(rpois(1000000, n)))
>  [1]  0.997961  1.999509  3.000748  3.999825  5.002983  6.004424  6.998825
>  [8]  7.997999  8.998341  9.766942 10.781751 11.817181 12.856154 13.900369
> [15] 14.999378
> and again:
> > sapply(1:15, function(n) mean(rpois(1000000, n)))
>  [1]  0.998570  2.000155  2.999534  4.002987  4.997999  5.996326  6.996761
>  [8]  7.996728  8.997265  9.764972 10.793913 11.822874 12.859886 13.903846
> [15] 15.000311
> 
> Weird....?
> 
> Best,
> 
> Kjetil
> -- 
> Kjetil Kjernsmo
> Graduate astronomy-student                    Problems worthy of attack
> University of Oslo, Norway            Prove their worth by hitting back
> E-mail: kjetikj at astro.uio.no                                - Piet Hein
> Homepage <URL:http://www.astro.uio.no/~kjetikj/>
> Webmaster at skepsis.no
> 
> 
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
>
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
----------------------
Alan T. Arnholt
Associate Professor
Dept. of Mathematical Sciences
TEL: (828) 262-2863
FAX: (828) 265-8617
arnholt at math.appstate.edu

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Anders Nielsen

2000-Nov-08 16:02 UTC

head link

[R] Strange means of numbers drawn from rpois

I can add:

Anders

> version; mean(rpois(100000, 15))         _         
platform Windows   
arch     x86       
os       Win32     
system   x86, Win32
status             
major    0         
minor    90.0      
year     1999      
month    November  
day      22        
language R         
[1] 14.98694



> version; mean(rpois(100000, 15))         _         
platform Windows   
arch     x86       
os       Win32     
system   x86, Win32
status   Patched   
major    0         
minor    99.0      
year     2000      
month    February  
day      9         
language R         
[1] 14.99506


> version; mean(rpois(100000, 15))         _         
platform Windows   
arch     x86       
os       Win32     
system   x86, Win32
status             
major    1         
minor    0.0       
year     2000      
month    February  
day      29        
language R         
[1] 14.99713

> version; mean(rpois(100000, 15))         _         
platform Windows   
arch     x86       
os       Win32     
system   x86, Win32
status             
major    1         
minor    1.0       
year     2000      
month    June      
day      15        
language R         
[1] 14.48548


> version; mean(rpois(100000, 15))         _         
platform Windows   
arch     x86       
os       Win32     
system   x86, Win32
status             
major    1         
minor    1.1       
year     2000      
month    August    
day      15        
language R         
[1] 14.50388



-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

ben@zoo.ufl.edu

2000-Nov-08 16:06 UTC

head link

[R] Strange means of numbers drawn from rpois

Happens on linux RH 6.0 R-1.1.1 as well, and furthermore a brief poke
into src/nmath/rpois.c shows that it's using a different method for
lambda>=10, so there's reason to believe there's a fairly localized
(but
scary!!!) bug.  I can't really understand/find/fix it, without a lot more
work and looking up the original reference (the rejection method given in
_Numerical Recipes_ is quite different) ...

 I'd say the following is a pretty clear indication of a bug ...
> mean(rpois(100000,9.99))
[1] 10.00021> mean(rpois(100000,10.0))[1] 9.50424


-- 
318 Carr Hall                                bolker at zoo.ufl.edu
Zoology Department, University of Florida    http://www.zoo.ufl.edu/bolker
Box 118525                                   (ph)  352-392-5697
Gainesville, FL 32611-8525                   (fax) 352-392-3704

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Possibly Parallel Threads

Search for more maybe matching threads

R help - Nov 2000 - Strange means of numbers drawn from rpois

[R] Strange means of numbers drawn from rpois

[R] Strange means of numbers drawn from rpois

[R] Strange means of numbers drawn from rpois

[R] Strange means of numbers drawn from rpois

Possibly Parallel Threads