Question

我是R的初学者，希望得到专家的帮助。我想创建一个函数来计算3个条件下的风险

1st：control：Xhh = 0 Xmi = 0

2st：hh：Xhh = 1 Xmi = 0

第3：hh + mi：Xhh = 1 Xmi = 1

并比较两组

组1：Xenv = 50

第2组：Xenv = 90

我的参数：

thi     lambda1 lambda2 lambda3 Beta Z2     Z1      Z4     Z3    Z6    Z5    theta
1.38    0.34    0.25    0.49    0.5  0.58   0.55    0.59   0.56  0.44  0.61  0.88

我想将所有这些参数插入此等式中 http://i.stack.imgur.com/DYR81.png

并计算不同时间点的值， Ti = 1至10

对于Ti = 0，将值设为0

然后用不同时间点的值绘制图形，用这三个条件的3条曲线进行比较，并在两组之间进行比较。所以最后在图中有6条曲线。

有人可以提供一些帮助吗？

Answer 1

这可能是对你的需求的一点解释，而不是你倾向于到达这里，但我正在寻找一个拖延今天早上其他任务的理由。

我认为这符合您的要求：

#define a function called myfunction
myfunction <- function(lambda1 = 0.34
                       ,lambda2 = 0.25
                       ,lambda3 = 0.49
                       ,Beta = 0.5
                       ,Z2 = 0.58
                       ,Z1 = 0.55
                       ,Z4 = 0.59
                       ,Z3 = 0.56
                       ,Z6 = 0.44
                       ,Z5 = 0.61
                       ,theta = 0.88
                       ,Xenv
                       ,Xhi
                       ,Xmi
                       ,Tmin
                       ,Tmax){
  #this should make this function somewhat generalizable to values of T
  #create an empty vector to hold values of our function as defined
  f <- rep(NA, length(Tmin:Tmax))
  #loop through values of T in your function
  #check parentheses here - I make no promises
  #I'm also unclear what your value of thi is. I may be missing something, 
  #but I don't see it in the function you have written
  i <- Tmin:Tmax
  f <- -log(exp((-(lambda1*i)^theta)*exp(log10(1-Beta*Xhi)+log10((Xenv/100)*(Z2-Z1)+Z2))-
                  ((lambda2*i)^theta)*exp(log10(1-Beta*Xmi))*log10((Xenv/100)*(Z4-Z3)+Z4)-
                  ((lambda3*i)^theta)*exp(log10((Xenv/100)*(Z6-Z5)+Z6)))*(1-theta)+theta)     
  #set f=0 at T=0 (I think this is what you want)
  if(Tmin==0) f[1] <- 0
  return(f)
}


#you didn't specify how to plot, but this seems to lend itself to a ggplot facted viz.
require(ggplot2)
require(reshape2)

#calculate for group 1
datg1 <- data.frame(t = 0:10
                    ,group = 1
                    ,condition1 = myfunction(Xenv=50, Xhi=0, Xmi=0, Tmin=0, Tmax=10)
                    ,condition2 = myfunction(Xenv=50, Xhi=1, Xmi=0, Tmin=0, Tmax=10)
                    ,condition3 = myfunction(Xenv=50, Xhi=1, Xmi=1, Tmin=0, Tmax=10)
)

#calculate for group 2
datg2 <- data.frame(t = 0:10
                    ,group = 2
                    ,condition1 = myfunction(Xenv=90, Xhi=0, Xmi=0, Tmin=0, Tmax=10)
                    ,condition2 = myfunction(Xenv=90, Xhi=1, Xmi=0, Tmin=0, Tmax=10)
                    ,condition3 = myfunction(Xenv=90, Xhi=1, Xmi=1, Tmin=0, Tmax=10)
)

#bind values together
dat <- rbind(datg1, datg2)

#melt your data into long format
datm <- melt(dat, id.vars = c("t", "group"))

#plot and facet
ggplot(datm, aes(x=t, y=value, colour=variable)) + 
  geom_line() + 
  facet_grid(.~group)

enter image description here

R函数用于计算不同时间点的值并绘制图形

1 个答案: