[Hps-analysis] [EXTERNAL] Profile error bars

[Nelson, Timothy Knight]

Hi All,

It appears that the RMS in z0 is coming out much larger than the apparent width of the distribution.  There are some outliers, so perhaps there is a flat piece that is so broad it ruins the RMS as an estimator, which seems notable in itself.  This is also apparent in some bins that have inexplicably large error bars.  In any case, this won't change the fitted slopes much, but rather just inflate the errors by a factor (5?)

Going back to the half dozen or so fabrication drawings needed to build up the wire positions in the nominal reference frame seems likely to produce the same result it has on the previous two passes.  Another route would be to have Bob Miller open the solid model (possible I could get someone at SLAC also) and measure directly.  That would cross check against any error I might be making systematically in adding (and subtracting) up all the factors.

Tim

> On Mar 21, 2023, at 10:08 AM, Nelson, Timothy Knight via Hps-analysis <hps-analysis at jlab.org> wrote:
> 
> Hello Norman,
> 
> I was curious about the slope errors on your z0 vs. tanL fits: they seem awfully large.  It looks to me like this originates with the error bars in the profile plots which are clearly much larger than one sigma in the width of the z0 distribution in each bin.  I wonder what can be going on here?  
> 
> Best,
> Tim
> 
> 
>  <page3image264685584.png>
> _______________________________________________
> Hps-analysis mailing list
> Hps-analysis at jlab.org
> https://mailman.jlab.org/mailman/listinfo/hps-analysis