[R] MANOVA

[David Winsemius]

Another method would be to use a summary that incorporates both as a  
measure of obesity, In medical investigations it is common to use the  
BMI which is the ratio of (weight in Kg) to (height in meters squared).

Yet a third method would be to investigate for nonlinearity on the  
response function using splines in the model. I have not yet seen  
evidence of collinearity offered,  so creating the full model as Simon  
suggests might be a first step. Much will depend on the quantity of  
data. If you only have 100 observations there will be severe  
limitations on the options.

-- 
David Winsemius
On Mar 12, 2009, at 2:37 AM, Simon Blomberg wrote:

> You only have one response variable, so MANOVA is not appropriate. One
> option would be to compare BP ~ Weight + Height with BP ~ 1. That  
> would
> give you a joint test of weight and height together. Since they are
> collinear, that should tell you the overall effect of "size". There  
> are
> other options, most of which involve discarding some of the data.  
> Frank
> Harrell's book is a font of wisdom on this sort of thing.
>
> Harrell, F. E., Jr. (2001). Regression Modeling Strategies. Springer.
>
> Simon.
>
> On Thu, 2009-03-12 at 00:20 -0600, Ding Xiao wrote:
>> Hi All,
>>
>> I have questions about MANOVA which I am still not sure if  
>> appropriately I should use it.
>>
>> For example I have a data set like this:
>>
>> BloodPressure (BP)  Weight   Height
>> 120        115    165
>> 125        145    198
>> 156        99      176
>>
>> I know that BloodPressure is correlated with both Weight and  
>> Height, however colinearity exists between Weight and Height. When  
>> I use BP = Weight + Height as the model, one is got to be  
>> insignificant. I was trying to use a BP + Weight = Height model,  
>> but not sure how to use it.
>>
>> Should I use MANOVA? or I just have to do two equations as BP =  
>> Weight & Weight = Height
>>
>> Any suggestions and answers are greatly appreciated!
>>
>> Ding
>>
>> 	[[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.
> -- 
> Simon Blomberg, BSc (Hons), PhD, MAppStat.
> Lecturer and Consultant Statistician
> School of Biological Sciences
> The University of Queensland
> St. Lucia Queensland 4072
> Australia
> Room 320 Goddard Building (8)
> T: +61 7 3365 2506
> http://www.uq.edu.au/~uqsblomb
> email: S.Blomberg1_at_uq.edu.au
>
> Policies:
> 1.  I will NOT analyse your data for you.
> 2.  Your deadline is your problem.
>
> The combination of some data and an aching desire for
> an answer does not ensure that a reasonable answer can
> be extracted from a given body of data. - John Tukey.
>
> ______________________________________________
> 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.

David Winsemius, MD
Heritage Laboratories
West Hartford, CT