I am probably a too bad mathematician and programmer, to buid a procedure for that ;) Here 2 Links to Origin. To be honest, no idea what's the calculation behind. I allways thought it would have somthing to do with Gausian Error Calc - but was just a guess. http://www.originlab.com/ http://en.wikipedia.org/wiki/Origin_(software) But I am working on a Mac, and Origin is only a PC tool. :)

I am probably a too bad mathematician and programmer, to buid a procedure for that ;) Here 2 Links to Origin. To be honest, no idea what's the calculation behind. I allways thought it would have somthing to do with Gausian Error Calc - but was just a guess. http://www.originlab.com/ http://en.wikipedia.org/wiki/Origin_(software) But I am working on a Mac, and Origin is only a PC tool. :)

hey there, and thx for answer :) I am not 100% sure if i did get what you have done, but i think it is not exactly what i meaned. The 2nd version you posted, where you used the ScatterPlot function, is how i have done it so far. But as the ScatterPlot function also has an option for X&Y Errorbars, i didn't additionally use that ErrorPlot function. So this i am allready able to do. What i meaned was that: • The Fit or NonlinearFit command, don't seem to take any care on the X&Y errors of my Datapoints. This functions are only based on the Datapoints. So the Fitted curve would alsways look the same, if i have Errorbars as tiny as possible, or as large as an elephant :). I want the Fitfunction to involve the errors in the calculation. • 2nd thing i would want to do is: Once my fit is calculated and ploted, and i read out any point on this Graph, i would like to read out the value + the error. So as an example: I am fitting a straight line: y = kx + d to my points. Then i want to know k. It is no problem to determine k, but i would also like to get the information about the error of k. And this error should be based on the errors of the Datapoints i entered. If it is not clear what i mean, plz say it, then i will try to put a scatch online somewhere, to visualize what i mean ;). Most of my friends work with a software called Origin to plot their Graphs with Errocalculation. Origin is able to take care of all that. I hoped there is a way to do that in Maple too, what i am sure it is, but i didn't succeed so far :).

hey there, and thx for answer :) I am not 100% sure if i did get what you have done, but i think it is not exactly what i meaned. The 2nd version you posted, where you used the ScatterPlot function, is how i have done it so far. But as the ScatterPlot function also has an option for X&Y Errorbars, i didn't additionally use that ErrorPlot function. So this i am allready able to do. What i meaned was that: • The Fit or NonlinearFit command, don't seem to take any care on the X&Y errors of my Datapoints. This functions are only based on the Datapoints. So the Fitted curve would alsways look the same, if i have Errorbars as tiny as possible, or as large as an elephant :). I want the Fitfunction to involve the errors in the calculation. • 2nd thing i would want to do is: Once my fit is calculated and ploted, and i read out any point on this Graph, i would like to read out the value + the error. So as an example: I am fitting a straight line: y = kx + d to my points. Then i want to know k. It is no problem to determine k, but i would also like to get the information about the error of k. And this error should be based on the errors of the Datapoints i entered. If it is not clear what i mean, plz say it, then i will try to put a scatch online somewhere, to visualize what i mean ;). Most of my friends work with a software called Origin to plot their Graphs with Errocalculation. Origin is able to take care of all that. I hoped there is a way to do that in Maple too, what i am sure it is, but i didn't succeed so far :).

thx as well :) will have a look at that

thx as well :) will have a look at that

hey there, Thx a lot! I'll try that out. I found out another solution this morning myself: It also works with a sequence, as in: F: = < seq( f( X[i] ), i=1..xmax ) > But the map-Method you described would be even better, because then i don't even would have to count the Number of entries, and don't need an index variable :). So will try that out as well. thx again, greets Gernot

hey there, Thx a lot! I'll try that out. I found out another solution this morning myself: It also works with a sequence, as in: F: = < seq( f( X[i] ), i=1..xmax ) > But the map-Method you described would be even better, because then i don't even would have to count the Number of entries, and don't need an index variable :). So will try that out as well. thx again, greets Gernot