Abstract
Programs for the HP-67/97 and Ti-59 programmable calculators have been written to fit data to the normal and log-normal distribution functions using linear regression. Each program calculates the value of the error function, erf, and the inverse error function, erf −1, in fitting the distribution functions. The erf is calculated using its series expansion. The erf −1 is calculated by successive approximations using generated values of the erf and the Newton-Raphson method in conjunction with an initial guessing function to accelerate convergence. The coefficient of determination is calculated in order to determine which distribution best fits the data. Both the mean and standard deviation of the distribution function can be determined from the linear regression; therefore the program is useful in several areas, such as settler design, tracer data analysis and particulate surface area determinations.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call