Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-L] re Bayesian Inference in Half-Life measurement





On 2021/Sep/24, at 06:40, Brian Whatcott <betwys1@sbcglobal.net> wrote:

Bernard, you mentioned using MCS at school. But how did one identify the decay constants of two isotopes using a geiger count times series in your old school days?
I see that Matlab offers potted examples of solving such conundrums with Bayesian models and Monte Carlo methods of evaluating hundreds or thousands of 'slices'.
On Thursday, September 23, 2021, 06:41:53 PM CDT, bernard cleyet <bernard@cleyet.org <mailto:bernard@cleyet.org>> wrote:


Brian!

Old school is sixty years ago. So, all I remember is graphing (Too old for most of you-all? It's W/paper and pencil). After the short one has “essentially died out” one draws a line W/straight edge averaging by eye. [1] then subtracting. More recently, was using the neutron source, previously mentioned. (ca. 1985=> ’89) However, having incipient Alzheimer’s, I don’t remember the detail, except Unix programs (before the word app.?) was used. linefit, etc., (no cursor and mouse) I’m rather certain I have a copy of that experiment by one of my student workers. If very interested, I will “hunt it up”.

[1] reminds me of “Numerical Recipes”, “Those who do Chi by eye, get what they deserve”.


bc