Question

我已经编写了一个UDF，其中我尝试连续运行4个while循环。它们是单独的循环，因为它们各自使用稍微不同的输入，但是当执行函数时，只有第一个while循环的结果存放在数据框中，我无法弄清楚原因。

我的代码在下面，应该是可执行的：

# polydata is a set of specific numbers, but here a random sample is fine.
polydata = sample(1:60, 91)

# User Defined Function - argument "f.day" is an integer, I have been testing with "5"
My.Function = function(f.day) {
  # Fit polynomial
  forecast.day = f.day
  x = seq(forecast.day, forecast.day + 90, 1)
  y.2 = coef(lm(polydata ~ x + I(x^2)))
  y.3 = coef(lm(polydata ~ x + I(x^2)+(x^3)))
  y.4 = coef(lm(polydata ~ x + I(x^2)+(x^3)+(x^4)))
  y.5 = coef(lm(polydata ~ x + I(x^2)+(x^3)+(x^4)+(x^5)))


  df = as.data.frame(matrix(NA, nrow = forecast.day+1, ncol = 4))
  colnames(df) = c("Quadratic", "Cubic", "Quartic", "Quintic")

  degree = 2
  i = 0
  df$Quadratic = c(rep(y.2[1], times = forecast.day + 1))
  while(i <= forecast.day + 1){
    j = 1
    while(j <= degree){
      df$Quadratic[i] = df$Quadratic[i] + y.2[j + 1] * i ^ j
      j = j + 1
    }
    i = i + 1
  }

  degree = 3
  i = 0
  df$Cubic = c(rep(y.3[1], times = forecast.day + 1))
  while(i <= forecast.day + 1){
    j = 1
    while(j <= degree){
      df$Cubic[i] = df$Cubic[i] + y.3[j + 1] * i ^ j
      j = j + 1
    }
    i = i + 1
  }

  degree = 4
  i = 0
  df$Quartic = c(rep(y.4[1], times = forecast.day + 1))
  while(i <= forecast.day + 1){
    j = 1
    while(j <= degree){
      df$Quartic[i] = df$Quartic[i] + y.4[j + 1] * i ^ j
      j = j + 1
    }
    i = i + 1
  }

  degree = 5
  i = 0
  df$Quintic = c(rep(y.5[1], times = forecast.day + 1))
  while(i <= forecast.day + 1){
    j = 1
    while(j <= degree){
      df$Quintic[i] = df$Quintic[i] + y.5[j + 1] * i ^ j
      j = j + 1
    }
    i = i + 1
  }
  return(df)
}

当我在测试脚本中运行该函数时，似乎只有第一个While循环正在完成：

> My.Frame = My.Function(5)
> My.Frame
  Quadratic Cubic Quartic Quintic
1  39.43232    NA      NA      NA
2  38.54225    NA      NA      NA
3  37.66369    NA      NA      NA
4  36.79664    NA      NA      NA
5  35.94109    NA      NA      NA
6  35.09705    NA      NA      NA

是否缺少某些内容或是否无法在函数内运行连续循环？

谢谢！

Answer 1

连续While循环可以在函数内运行，就像在普通脚本中一样。这里的错误是生成y.2, y.3, y.4, y.5向量。这里系数被生成到特定的程度-2（二次），3（立方），4（四次）和5（五次）。

y.2是正确生成的，但其余部分缺少必要数量的I元素，因此只为它们全部生成3个系数（这是仅二次方的正确数量）。因此，当j = 3（度）时，那么在df$Cubic[i] = df$Cubic[i] + y.3[j + 1] * i ^ j中，它正在索引y.3 [4]，这将给出NA，因为它不存在。

原始代码如下：

  y.2 = coef(lm(polydata ~ x + I(x^2)))
  y.3 = coef(lm(polydata ~ x + I(x^2)+(x^3)))
  y.4 = coef(lm(polydata ~ x + I(x^2)+(x^3)+(x^4)))
  y.5 = coef(lm(polydata ~ x + I(x^2)+(x^3)+(x^4)+(x^5)))

读作：

  y.2 = coef(lm(polydata ~ x + I(x^2)))
  y.3 = coef(lm(polydata ~ x + I(x^2)+I(x^3)))
  y.4 = coef(lm(polydata ~ x + I(x^2)+I(x^3)+I(x^4)))
  y.5 = coef(lm(polydata ~ x + I(x^2)+I(x^3)+I(x^4)+I(x^5)))

因此，当现在运行时，将根据请求的程度产生正确数量的系数，整个函数现在将产生完整的数据框。

你可以在R中的用户定义函数内运行连续的While循环吗？

1 个答案: