[R] Relation 1.5*IQR/Percentile in case of a normal Distribution

Peter Dalgaard pdalgd at gmail.com
Sat Aug 14 23:08:31 CEST 2010


Johannes Graumann wrote:
> Hi,
> 
> can someone point me at material to understand how in 
> http://upload.wikimedia.org/wikipedia/commons/8/89/Boxplot_vs_PDF.png the 
> "fivenum"-corresponding percentages might be calculated?

Looks like a pretty straightforward application of pnorm() and qnorm().

pnorm(4*qnorm(.75), lower=F) # Q3 + 1.5 IQR = 4 Q3 since IQR = 2 Q3

[1] 0.003488302

gives the tail probabilities of .35%, and the rest is by definition.

-- 
Peter Dalgaard
Center for Statistics, Copenhagen Business School
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com



More information about the R-help mailing list