Re: [Phys-L] Half-Life measurement

[Brian Whatcott <betwys1@sbcglobal.net>]

 Paul, you mentioned this:
one of the model fit numbers came within 0.5% of the world
value.  While that is nice to see, I should remind you that the standard
deviation on that fit was still ~10%.  
Which reminds me; you mentioned a half life value on a second data set was 1.5 SDsoff the published value. What percentage difference was that?
    On Friday, October 15, 2021, 05:50:37 PM CDT, Paul Nord <paul.nord@valpo.edu> wrote:  
 
 Brian,

You note that one of the model fit numbers came within 0.5% of the world
value.  While that is nice to see, I should remind you that the standard
deviation on that fit was still ~10%.  There are infinitely many incorrect
solutions that will still give you the answer in the back of the book.  The
number really isn't as good as it seems to be.

Paul


On Fri, Oct 15, 2021 at 2:59 PM Brian Whatcott <betwys1@sbcglobal.net>
wrote:

> John Denker via Phys-l <phys-l@mail.phys-l.org
> > UnsubscribeTo:phys-l@phys-l.orgCc:John Denker
> Thu, Oct 14 at 7:56 PM
> Addendum to previous message:Here is what the fits look like:
> https://www.av8n.com/copper-decay/img48/rate-v-short-time-1.png
> https://www.av8n.com/copper-decay/img48/rate-v-long-time-1.pngor
> equivalently
> https://www.av8n.com/copper-decay/pdf/rate-v-short-time-1.pdf
> https://www.av8n.com/copper-decay/pdf/rate-v-long-time-1.pdf
> This is pretty much what you expect to see when fitting tonoisy data.
>
> ###############################################################################
> The two plots John gave above were plotted from data in his previous
> message. This mentioned two datasets: "my_data" and "sean_data". I am not
> sure which data developed the plots. The iterative quadratic approximation
> method he used in lopt provided the following estimates.5.587 min for the
> shorter half life + 9% high  )13.68 hrs for the longer half life  + 7.7%
> high) Background rate was 12.7 counts/min          )  using  my_data (not
> sure if he incorporated Pauls corrected values around 3 minutes)
> 5.264 min for the shorter half life was +2.8% high )13.67 hrs for the
> longer half life was 12.1% high  ) Background rate was  14.35 counts/min
>             ) using Sean_data
> This nlopt routine does not appear to use Bayesian probability
> estimates, but I think John mentioned using published values of half
> lives for his initial iteration.( what I called brute force Bayesian in my
> estimates)
> These compare well with Paul Nord's Bayesian estimates below:
> 4.67 min for the shorter half life was 8.8% low        )12.76 hrs for the
> longer half life was 0.5% high (!)  )Background rate was 13.995
> counts/min                ) using my_data
>
> So far so good! Finally,  I admit to some puzzlement about the plots which
> John provided.
> When I digitized his plots I found the Short tau was 8.74 min, 6.06 min
> half life which is 18% highand the long tau was 8.05 hrs, half life 5.58
> hrs was extremely low at -56%The background rate also seemed very high.
>   Not sure how I could stray so far from his good data.
> _______________________________________________
> Forum for Physics Educators
> Phys-l@mail.phys-l.org
> https://www.phys-l.org/mailman/listinfo/phys-l
_______________________________________________
Forum for Physics Educators
Phys-l@mail.phys-l.org
https://www.phys-l.org/mailman/listinfo/phys-l