Bennett5_opt: Magentization modelling

In NISTopt: Nonlinear least squares examples from NIST in form of functions for optim() and optimx()

Description

Data and functions for fitting the Bennett5 nonlinear least squares problem. The Bennett5 data frame has 154 rows and 2 columns of data from a magnetism study

Usage

Bennett5.f(b)

Arguments

b	A numeric vector of starting estimates.

Format

This data frame contains the following columns:

y

A numeric vector of magnetism values.

x

A numeric vector of log(time).

Details

These data are the result of a NIST study involving superconductivity magnetization modeling. The response variable is magnetism, and the predictor variable is the log of time in minutes.

Value

The numeric value of the sum of squared deviations between model and data.

Source

Bennett, L., L. Swartzendruber, and H. Brown, NIST (1994). Superconductivity Magnetization Modeling.

Examples

Try <- function(expr) if (!inherits(val <- try(expr), "try-error")) val
plot(y ~ x, data = NISTopt:::Bennett5)
Try(fm1 <- nls(y ~ b1*(b2+x)**(-1/b3), data = NISTopt:::Bennett5,
           start = c(b1 = -2000, b2 = 50, b3 = 0.8), trace = TRUE))
Try(fm1a <- nls(y ~ b1*(b2+x)**(-1/b3), data = NISTopt:::Bennett5,
            start = c(b1 = -2000, b2 = 50, b3 = 0.8),
            trace = TRUE, alg = "port"))
Try(fm2 <- nls(y ~ b1*(b2+x)**(-1/b3), data = NISTopt:::Bennett5,
           start = c(b1 = -1500, b2 = 45, b3 = 0.85), trace = TRUE))
Try(fm2a <- nls(y ~ b1*(b2+x)**(-1/b3), data = NISTopt:::Bennett5,
            start = c(b1 = -1500, b2 = 45, b3 = 0.85),
            trace = TRUE, alg = "port"))
Try(fm3 <- nls(y ~ (b2+x)**(-1/b3), data = NISTopt:::Bennett5, alg = "plinear",
           start = c( b2 = 50, b3 = 0.8), trace = TRUE))
Try(fm4 <- nls(y ~ (b2+x)**(-1/b3), data = NISTopt:::Bennett5, alg = "plinear",
           start = c( b2 = 45, b3 = 0.8), trace = TRUE))

NISTopt documentation built on July 25, 2017, 3:01 p.m.