[rkward-devel] failing tests

[meik michalke]

hi,

am Montag 24 September 2012, 19:07:38 schrieb Thomas Friedrichsmeier:
> On Friday 21 September 2012, meik michalke wrote:
> > <import_spss.messages.txt>
> > <import_stata.messages.txt>
> 
> Both work fine, here. At least for SPSS it looks like the test file could
> not be read at all:

it can't, but i still don't know why. read.spss() doesn't agree with it:

<snip>
> data <- read.spss ("import_export_plugins_testfile.sav", to.data.frame=TRUE, 
max.value.labels=1)
Error in read.spss("import_export_plugins_testfile.sav", to.data.frame = TRUE,  
: 
  error reading system-file header
Calls: read.spss -> .Call
In addition: Warning message:
In read.spss("import_export_plugins_testfile.sav", to.data.frame = TRUE,  :
  import_export_plugins_testfile.sav: position 1: Variable name begins with 
invalid character
Error in read.spss("import_export_plugins_testfile.sav", to.data.frame = TRUE,  
: 
  error reading system-file header
Calls: read.spss -> .Call
</snip>

the same thing happens in a native R shell, so this is not RKWard specific. 
looks more like a problem with the foreign package to me (0.8-50).

am Donnerstag 27 September 2012, 11:38:19 schrieb Thomas Friedrichsmeier:
> On Friday 21 September 2012, meik michalke wrote:
> >  - NOTE: ad_test, cvm_test, pearson_test, sf_test, lillie_test and
> >    pareto_chart were skipped for now (missing libs)
> 
> these work for me (but of course, it would be great, if you could install
> the missing libs and re-run the tests to confirm).

ok, did that. they all pass, except for ad_test, which is mildly off the 
standards because of differences in rounding precision:

<snip>
[2] "< <tr><td>perm</td><td>48</td><td>0</td><td>A = 
3.9918437155579</td><td>4.2977e-10</td></tr>" 
[3] "---"                                                                                          
[4] "> <tr><td>perm</td><td>48</td><td>0</td><td>A = 
3.99184371555789</td><td>4.2977e-10</td></tr>"
</snip>

> >  - kpss_test and anscombe_test failed here because the results were
> > slightly off the standards (looks plausible, but should be confirmed by
> > someone else)
> 
> I've changed anscombe_test to be fuzzy (and thus to pass). kpss_test matches
> the standard, here. How far off are the numbers for you?

<snip>
[2] "< 
<tr><td>test50y</td><td>50</td><td>0</td><td>0.12985</td><td>5</td><td>0.079915</td></tr>"
[3] "< 
<tr><td>test50z</td><td>50</td><td>0</td><td>0.071534</td><td>5</td><td>0.1</td></tr>"    
[4] "---"                                                                                        
[5] "> 
<tr><td>test50y</td><td>50</td><td>0</td><td>0.1378</td><td>5</td><td>0.065181</td></tr>" 
[6] "> 
<tr><td>test50z</td><td>50</td><td>0</td><td>0.099722</td><td>5</td><td>0.1</td></tr>"    
</snip>


viele grüße :: m.eik

-- 
dipl. psych. meik michalke
institut f"ur experimentelle psychologie
abt. f"ur diagnostik und differentielle psychologie
heinrich-heine-universit"at d-40204 d"usseldorf
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2282 bytes
Desc: not available
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20120930/d202c3f1/attachment.p7s>