Re: st: BatPlot options and Creating a Matrix

[Nick Cox <njcoxstata@gmail.com>]

On the second question, check out -statsby-.

On the first, see also -concord- (-findit concord- for locations).

Nick

On 4 Aug 2011, at 12:44, Diego Bellavia <messadua@yahoo.it> wrote:

> Dear Statalisters,
>
> I am struggling with batplot (quiet successfully I must say) and I  
> have a couple
> of questions regarding the use of this command:
>
>
> 1) Is there any way to modify the Y axis label options so that the  
> "dashed"
> green line would be centered to 0 value on Y axis and the limits of  
> agreements
> would be symmetric around the zero then ? Using the ylab options I  
> can change
> the label but not the position of the graph...
>
> 2) The "grey" area defined by the B-A plot (using the notrend  
> option) defines
> 95% limits of agreeement, am I correct ?
>
>
> 3) My Mentor here asked me to compute the "Bias" for the inter- 
> observer
> agreement in the same dataset of measures. I am assuming this is the  
> simple
> coefficient of variation (obtained as 100 * SDdiff / Mean-diff) am I  
> correct ?
>
>
> I have also another question, totally unrelated to B-A plots:
>
> I have to run several simple regression models and would like to  
> automatically
> group results
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/