[R] Binomial PCA Using pcr()

Ulrik Stervbo U|r|k@Stervbo @end|ng |rom ruhr-un|-bochum@de
Wed Aug 19 10:52:57 CEST 2020


Hi Prasad,

I think this might be a problem with the package, and you can try to 
contact the package author.

The error seem to arise because the pcr() cannot find the 
'negative-binomial' distribution

```
library(qualityTools)
x <- rnbinom(500, mu = 4, size = 100)
pcr(x, distribution = "negative-binomial")
```

When I look in the code of pcr(), there is some testing against the 
words 'negative binomial' (note the missing -), although the 
documentation clearly lists 'negative-binomial' as a possible 
distribution.

Unfortunately changing 'negative-binomial' to 'negative binomial' does 
not help, as

```
pcr(x, distribution = "negative binomial")
```

throws the error "object '.confintnbinom' not found" and a lot of 
warnings.

Best,
Ulrik


On 2020-08-12 12:50, Prasad DN wrote:
> Hi All,
> 
> i am very new to R and need guidance.
> 
> Need help in doing process capability Analysis for my data set (6 
> months of
> data) given in below format:
> 
> Date   |   Opportunities  |  Defectives | DefectivesInPercent
> 
> I searched and found that pcr() from QualityTools package can be used 
> for
> this purpose.  The USL is 2% defectives.
> 
> MyData = read.csv(file.choose())   #select  CSV file that has data in 
> above
> mentioned format.
> x <- MyData$DefectivesInPercent
> 
> pcr(x, distribution = "negative-binomial", usl=0.02)
> 
> I get error message as:
> Error in pcr(x, distribution = "negative-binomial", usl = 0.02) :
>   y distribution could not be found!
> 
> Please advise, how to proceed?
> 
> Regards,
> Prasad DN
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide 
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list