Hello all,
I am estimating parameters for regression functions on experimental data.
Functional response of Rogers type II.
I would like to know which points of my dataset are outliers. What is the
best method to do this with R?
I found a method via R help, but would like to know if there are better
methods for my purpose.
Here is the script I us now:
library("mvoutlier")
dat <- read.delim("C:/data.txt")
uni.plot(dat)
My data looks like the following (copied into a txt file):
(N0 is the initial number of eggs fed to the predator, FR is the number of
eggs eaten by the predator during 24h)
N0 FR
37 30
27 15
36 14
37 13
45 8
25 0
47 20
34 6
25 8
21 7
24 24
34 17
23 10
29 5
38 38
24 24
20 17
14 8
18 15
15 10
26 5
33 5
22 21
38 3
22 20
23 19
20 6
20 4
21 18
25 5
13 13
9 8
8 4
7 7
8 5
11 9
Kind regards,
Met vriendelijke groeten,
Joachim
Don't waste paper! Think about the environment before printing this e-mail
______________________________________
Joachim Audenaert
Adviesdienst Gewasbescherming
Proefcentrum voor Sierteelt
Schaessestraat 18
B-9070 Destelbergen
Belgium
Tel. +32 9 353 94 71
Fax +32 9 353 94 95
E-mail: joachim.audenaert@pcsierteelt.be
www.pcsierteelt.be
______________________________________
[[alternative HTML version deleted]]
Hi, I believe that first learning the appropriate statistical methods to detect the outliers and searching for the related functions in R is a better way. Ozgur -- View this message in context: http://r.789695.n4.nabble.com/detecting-outliers-tp4632636p4632637.html Sent from the R help mailing list archive at Nabble.com.
Joachim Audenaert <Joachim.Audenaert <at> pcsierteelt.be> writes:> > Hello all, > > I am estimating parameters for regression functions on experimental data. > Functional response of Rogers type II. > > I would like to know which points of my dataset are outliers. What is the > best method to do this with R?The best method for detecting outliers really depends on the motivation/purpose. Your data look noisy, but by eye nothing really jumps out. Looking at a histogram (hist()) and Q-Q plot qqnorm() of the residuals of the fit, it looks like the distribution is slightly skewed but that there are no points that really fall very far outside a normal distribution (normality is not a necessity for making inferences from an nls fit, but it helps a lot)
Hello, Had you looked more, and you would have seen R-help discussions on what is an outlier. Almost unanimously, an ill defined concept. In your problem, predators don't eat all eggs that they are given except for one case, 38 were given and all 38 were eaten. You can detect this in R with boxplot.stats(d$FR) Or with the return value of boxplot. See ?boxplot Hope this helps, Rui Barradas Em 07-06-2012 07:24, Joachim Audenaert escreveu:> Hello all, > > I am estimating parameters for regression functions on experimental data. > Functional response of Rogers type II. > > I would like to know which points of my dataset are outliers. What is the > best method to do this with R? > I found a method via R help, but would like to know if there are better > methods for my purpose. > Here is the script I us now: > > library("mvoutlier") > dat <- read.delim("C:/data.txt") > uni.plot(dat) > > My data looks like the following (copied into a txt file): > (N0 is the initial number of eggs fed to the predator, FR is the number of > eggs eaten by the predator during 24h) > > N0 FR > 37 30 > 27 15 > 36 14 > 37 13 > 45 8 > 25 0 > 47 20 > 34 6 > 25 8 > 21 7 > 24 24 > 34 17 > 23 10 > 29 5 > 38 38 > 24 24 > 20 17 > 14 8 > 18 15 > 15 10 > 26 5 > 33 5 > 22 21 > 38 3 > 22 20 > 23 19 > 20 6 > 20 4 > 21 18 > 25 5 > 13 13 > 9 8 > 8 4 > 7 7 > 8 5 > 11 9 > > > Kind regards, > Met vriendelijke groeten, > Joachim > > Don't waste paper! Think about the environment before printing this e-mail > > ______________________________________ > > Joachim Audenaert > Adviesdienst Gewasbescherming > Proefcentrum voor Sierteelt > Schaessestraat 18 > B-9070 Destelbergen > Belgium > Tel. +32 9 353 94 71 > Fax +32 9 353 94 95 > E-mail: joachim.audenaert at pcsierteelt.be > www.pcsierteelt.be > ______________________________________ > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > 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. >