Skip to content

Continuous (Non-Count) Skewed Data With Many Zeros

3 messages · Highland Statistics Ltd, Rich Shepard, Mieke Zwart

#
Non-normality of your response variable is not a reason to apply a data 
transformation.
It all depends, and no sensible answer can be given. 15% of zeros can 
screw things up....but it is also possible that 80% of zeros comply with 
a regression or GLM. For a discussion with examples see Chapter 10 in 
our 2012 book.
Depends on the previous remark.....anything from linear regression to a 
zero inflated model for a continuous distributed response variable. 
There is just no simple answer possible. It all depends. But based on 
what you describe it will probably be something zero-inflated.

Alain
#
On Wed, 16 May 2012, Highland Statistics Ltd wrote:

            
Alain,

   I was exploring whether a transformation might be useful. Turns out that
it's not useful.
I recognize there's no one correct answer. I'm seeking options because
I've not before had a reason to dig deeply into data like this. I've ordered
your 2012 book[1] and expect it to arrive Real Soon Now.
Now I have a handfull of potential approaches to explore. I appreciate
your insights and comments as I delve into new areas.

Regards,

Rich
#
Hi Alain,

Just wondering if the 2012 book you are talking about is the "Zero Inflated Models and Generalized Linear Mixed Models with R". You say to look at chapter 10 but this book has only 9 chapters according to the website. Is it the book you are talking about?

Cheers,

Mieke

 
-----Original Message-----
From: r-sig-ecology-bounces at r-project.org [mailto:r-sig-ecology-bounces at r-project.org] On Behalf Of Highland Statistics Ltd
Sent: 16 May 2012 14:09
To: r-sig-ecology at r-project.org
Subject: Re: [R-sig-eco] Continuous (Non-Count) Skewed Data With Many Zeros
Non-normality of your response variable is not a reason to apply a data transformation.
It all depends, and no sensible answer can be given. 15% of zeros can screw things up....but it is also possible that 80% of zeros comply with a regression or GLM. For a discussion with examples see Chapter 10 in our 2012 book.
Depends on the previous remark.....anything from linear regression to a zero inflated model for a continuous distributed response variable. 
There is just no simple answer possible. It all depends. But based on what you describe it will probably be something zero-inflated.

Alain