Question

试图拟合三机制非线性膜污染模型，其包括积分表达式。试图从nls.lm {minpack.lm}帮助页面扩展示例1。问题：如何修改示例1以包含整数表达式？在一些搜索之后进行了各种尝试，这似乎是我能得到的最接近但仍然给我以下错误（“match.fun（f）中的错误：'exp（1）^（xx * parS $ b）'不是函数，字符或符号“）当然不会绘制曲线拟合。非常感谢任何帮助！

require(minpack.lm)

###### example 1 + 'integrate() function
## values over which to simulate data 
x <- seq(0,5,length=100)
xo <- x[1]

## two models based on a list of parameters
integrand <- function(parS=pp, xx=x) exp(1)^(xx * parS$b)
getPred <- function(parS, xx) parS$a * exp(xx * parS$b) + parS$c 
getPred2 = function(parS, xx) parS$a * exp(xx * parS$b)+ integrate(integrand, lower=xo, upper=xx) + parS$c

## parameter values used to simulate data
pp <- list(a=9,b=-1, c=6) 

## simulated data, with noise  
simDNoisy <- getPred(pp,x) + rnorm(length(x),sd=.1); simDNoisy
simDNoisy2 <- getPred2(pp,x) + rnorm(length(x),sd=.1); simDNoisy2

## plot data
plot(x,simDNoisy, main="data")

## residual functions
residFun <- function(p, observed, xx) observed - getPred(p,xx)
residFun2 <- function(p, observed, xx) observed - getPred2(p,xx)

## starting values for parameters  
parStart <- list(a=3,b=-.001, c=1)

## perform fit 
nls.out <- nls.lm(par=parStart, fn = residFun, observed = simDNoisy, xx = x)
nls.out2 <- nls.lm(par=parStart, fn = residFun2, observed = simDNoisy, xx = x)

## plot model evaluated at final parameter estimates  
lines(x, getPred(as.list(coef(nls.out)), x), col=2, lwd=2)
lines(x, getPred2(as.list(coef(nls.out2)), x), col=3, lty=2,lwd=2)

## summary information on parameter estimates
summary(nls.out) 
summary(nls.out2

）

NRussel建议修改后的代码并包含一个被积函数但仍然出错：“parS $ b：$运算符错误原子向量无效”试图找出原因。谢谢

How to pass vector to integrate function

Roland在链接1中提供的答案有助于确定如何对整体部分进行矢量化;但仍试图弄清楚如何将带参数的积分传递给nls.lm（）的预测函数。修改后的代码如下：

# vectorize integrand and plot of results
integrand <- function(parS, xx) exp(1)^(xx * parS$b)

res = c()
for(ii in 1:100){
  res[ii] = integrate(Vectorize(integrand, vectorize.args="xx"), lower=xo, upper=x[ii], parS=pp)
}
inres = unlist(res)
x
plot(x, inres)

非线性模型拟合积分 - R - 一般方法

0 个答案: