Rank: Member Groups: Registered
Joined: 08/06/2021(UTC) Posts: 16 Location: Караганда
|
Originally Posted by: Jean Giraud Normal PDF almost don't exist in physics unless intentional like shooting 100 times the target with the same gun. Under the hood there may be Logistic, Weibull noise. If you have these 3 data sets from raw experiment, please: attach as collected, absolutely untouched. Cheers ... Jean
So here it is. Two cases. I counted the first one in the very first examples. There, the three normal distributions are summed up. In the last example, the other case is that two normal distributions are summed up. But, actually, the problem is solved. Only the time for calculating each case is too long.
|
|
|
|
Rank: Member Groups: Registered
Joined: 08/06/2021(UTC) Posts: 16 Location: Караганда
|
Originally Posted by: Jean Giraud You forgot to attach the untouched experimental data set. That's the only thing I need in statistical work.
Why did I forget? Both sets of data are there. Here with two normal distributions (the third one is not visible). Here with three normal distributions. In principle, it is still necessary to take into account that these are histograms. This is given as X the middle of the column. Initially, the column width is 20 units, then 40 is for X=180, then in the first case the width is 100, and the last column is 50. In the second case, the last three columns are 50 each.
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 23/07/2013(UTC) Posts: 1,125 Was thanked: 506 time(s) in 337 post(s)
|
Originally Posted by: anatolsen Originally Posted by: Jean Giraud Problem is that we don't know what you are looking for. Are you looking for the best fit to your data set ??? If so, minute job. Please, don't hesitate ... Jean
Now only one question remains. Iterating through 9 parameters requires a lot of time. And the calculation is carried out with an accuracy that is superfluous for my purposes. Sq and Sqmin are calculated with an accuracy of 15 decimal places. And I would have more than enough 8 or 9 digits after the decimal point. This would significantly reduce the calculation time. And here is the Pearson correlation coefficient I have 0.9999, whereas I would have more than enough 0.9 Here is your Sq and Sqmin calculations. I have noticed Sqmin make a peak to lowest and rise again at some point. I coded to find those points range within first iteration. This is done by evaluating variables one by one, not all at one. All iterations are taken under 5 seconds on my PC. Please inform if this results are enough for you. Regards eng-min_rev_fast.pdf (618kb) downloaded 6 time(s). eng-min_rev_fast.sm (41kb) downloaded 4 time(s).Edited by user 15 June 2021 23:49:52(UTC)
| Reason: Not specified
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 23/07/2013(UTC) Posts: 1,125 Was thanked: 506 time(s) in 337 post(s)
|
A revised Sq and Sqmin calculations of previous. All iterations are taken under 25 seconds on my PC. Sqmin is dropped to 6x10^-6. And now I am out of ideas. Please inform if this results are enough for you. Regards eng-min_rev_fast.pdf (672kb) downloaded 10 time(s). eng-min_rev_fast.sm (48kb) downloaded 14 time(s).
|
|
|
|
Rank: Member Groups: Registered
Joined: 08/06/2021(UTC) Posts: 16 Location: Караганда
|
Colleagues
Thank you very much Apparently, we need to stop here - the calculation method has been worked out.
Once again-thank you very much
|
|
|
|
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.