0

Using the extremevalues package in R for univariate outlier detection. I have an admittedly limited understanding of statistics, but am trying to conceptually grasp what some of the getOutliersI function is doing to determine outliers. I've tried looking at the package's documentation and am not finding it.

Calling the function on my data produces results that fit with common sense:

getOutliersI(data)

Since I'm not supplying any rho or FLim arguments, It's calculating the limits for me--but how?

agstudy
  • 119,832
  • 17
  • 199
  • 261
taylorwc
  • 45
  • 1
  • 5

1 Answers1

0

I am not sure to understand your question. But a simple answer is if you don't supply argument it will use optional parameters.

getOutliersI(y, rho=c(1,1), FLim=c(0.1,0.9), distribution="normal")

So here it will use a normal distribution with parameter rho=c(1,1) and FLim=c(0.1,0.9 to compute Gaussian limits.

agstudy
  • 119,832
  • 17
  • 199
  • 261
  • Sorry for that ambiguous question; having trouble wording it... In my case, I'm not supplying any parameters, just the data. What are the default values it's using? – taylorwc Feb 11 '13 at 17:36
  • the defaults parametrs are `rho=c(1,1), FLim=c(0.1,0.9), distribution="normal"` – agstudy Feb 11 '13 at 17:38
  • Thanks. Maybe I'm just a noob, but the R documentation is nearly unintelligible. – taylorwc Feb 12 '13 at 17:53
  • @taylorwc really? maybe there is some learning curve but once you get the basics you can do many things. There many goods tutorials. Take a look a [this](http://stackoverflow.com/questions/1744861/how-to-learn-r-as-a-programming-language/1744882#1744882) for example. – agstudy Feb 12 '13 at 18:06