Some questions on simple statistical analysis

Hi All,

I would like to ask you some questions about simple statistical analysis.

In my hand, there are calibration constants and ledmean,pedmean,ledrms,pedrms.
So I can get these results ;

"truemean = ledmean-pedmean"
"truerms = sqrt(pow(ledrms,2)-pow(pedrms,2))"
"gain = truemean*2.6*pow(10,(-15))/(1.15*(pow(truemean,2)/pow(truerms,2)))*1.6*pow(10,(-19))"

Shortly,for simple analysis, ledmean of data is its actual value and pedmean of it is its error value
so we got it truemean = ledmean - pedmean, that is; true value!

To given information to me, “Relative Response = (ledmean-pedmean)/calibration_constants” and this is equal to
"truemean/calibration_constants"
I would like to ask you a question here!
Should i calculate “Relative Response truerms” with calibration_constants, that is;
relative truerms = sqrt(pow(ledrms,2)-pow(pedrms,2))/calibration_constants ? Does it make sense ?

And another question here is how am i able to calculate ‘relative response gain’ ?
it might as well this ;

relative response = truemean/calibration_constants
relative truerms = sqrt(pow(ledrms,2)-pow(pedrms,2))/calibration_constants
and finally relative gain is equal to  
relative response*2.6*pow(10,(-15))/(1.15*(pow(relative response,2)/pow(relative truerms,2)))*1.6*pow(10,(-19))"

Thanks for your help,
Cheers,
Ersel

Hi,

I am not sure I have understood your problem. It is not clear to me what your quantities are. For example why you define the true quantities like this ? Then what you call rms is the standard deviation or the root mean square ?
Since this questions are related more to your specific problematics and not to ROOT, I would suggest you to try to get an answer from your colleagues.

Best Regards

Lorenzo

Ok thanks Lorenzo. it will be better to do so…