Quantcast

Interpreting Q-Q Plots

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Interpreting Q-Q Plots

Rich Shepard
   My understanding of Q-Q plots is that if the tails of the plotted points
fall above or below the x=y line the distribution of observed/measured
values is under or over dispersed. But, how do I interpret measured values
that are in horizontal lines? The attached plot illustrates this situation.

TIA,

Rich
______________________________________________
[hidden email] 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.

chromium_norm.pdf (8K) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Interpreting Q-Q Plots

Peter Alspach-2
Tena koe Rich

Probably highly skewed to the right, with discrete values (perhaps due to the limitations in the accuracy of the assessment equipment).  But note:

library(fortunes)
fortune('chicken')

HTH .....

Peter Alspach

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf Of Rich Shepard
Sent: Tuesday, 15 May 2012 9:53 a.m.
To: [hidden email]
Subject: [R] Interpreting Q-Q Plots

   My understanding of Q-Q plots is that if the tails of the plotted points fall above or below the x=y line the distribution of observed/measured values is under or over dispersed. But, how do I interpret measured values that are in horizontal lines? The attached plot illustrates this situation.

TIA,

Rich

The contents of this e-mail are confidential and may be subject to legal privilege.
 If you are not the intended recipient you must not use, disseminate, distribute or
 reproduce all or any part of this e-mail or attachments.  If you have received this
 e-mail in error, please notify the sender and delete all material pertaining to this
 e-mail.  Any opinion or views expressed in this e-mail are those of the individual
 sender and may not represent those of The New Zealand Institute for Plant and
 Food Research Limited.

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Interpreting Q-Q Plots

Rich Shepard
On Tue, 15 May 2012, Peter Alspach wrote:

> Probably highly skewed to the right, with discrete values (perhaps due to
> the limitations in the accuracy of the assessment equipment).

Peter,

   Most of these data are near zero or the lower detection limit. A few
values are very much higher. I didn't think of skewness as a reason.

>  But note:
>
> library(fortunes)
> fortune('chicken')

   And since I don't have the experience, the only way to gain it is by
learning from those with practice reading chicken entrails.

Thanks,

Rich

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Interpreting Q-Q Plots

S Ellison-2
>    And since I don't have the experience, the only way to gain it is by
>  learning from those with practice reading chicken entrails.
This can be hard on the chicken population.

Try comparing QQ plots for simulated random data from different distributions with something more immediately interpretable on the measurement scale,  such as dot plots, box plots and density plots. That should add up to a fair bit of experience quite quickly.

#Example
par(mfrow=c(1,2))
qqnorm(x<-rlnorm(200, 1,0.5))
qqline(x)
plot(density(x))

qqnorm(x<-rnorm(200, sample(c(0,4), 200, replace=TRUE))) #bimodal
qqline(x)
plot(density(x))


and so on.

Notice that qqnorm's vertical scale by defult corresponds to the horizontal scale in density plots and stripcharts. I personally prefer datax=TRUE, but really that's only a choice about whether to face north or east when reading the entrails.

*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Interpreting Q-Q Plots

Keith Jewell
In reply to this post by Rich Shepard
On 14/05/2012 23:21, Rich Shepard wrote:

> On Tue, 15 May 2012, Peter Alspach wrote:
>
>> Probably highly skewed to the right, with discrete values (perhaps
>> due to
>> the limitations in the accuracy of the assessment equipment).
>
> Peter,
>
>   Most of these data are near zero or the lower detection limit. A few
> values are very much higher. I didn't think of skewness as a reason.
>
>>  But note:
>>
>> library(fortunes)
>> fortune('chicken')
>
>   And since I don't have the experience, the only way to gain it is by
> learning from those with practice reading chicken entrails.
>
> Thanks,
>
> Rich
>
I found this page helpful
http://www.cms.murdoch.edu.au/areas/maths/statsnotes/samplestats/qqplot.html

HTH

Keith J

______________________________________________
[hidden email] 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.
Loading...