Dear Colleagues, I have an equation Y^2 = (2D)X, where "2D" is a constant and both Y and X are experimentally-measured parameters. I have both experimental values of X (values known precisely; std. deviation = 0) and of Y (for which I have std. deviations "y" for each Y-value). To obtain the value of D, I plotted Y^2 vs. X and divided the resultant slope by 2. The value of D thus obtained corresponded to our expectations.
My question is: if I have 5 values of Y +/- y and 5 values of X (+/- 0), is there a way of calculating the std. deviation for the value of D (i.e., D +/- d)? Responders are kindly requested to contact me _directly_ at [EMAIL PROTECTED] because I don't often consult this usegroup. Please be as specific as possible in your reply, because I'm no mathematician. Thanks in advance to all responders, S. Shapiro [EMAIL PROTECTED] ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at http://jse.stat.ncsu.edu/ =================================================================