varianceBy<T> function Summarizing data

num? varianceBy<T>(Iterable<T> iterable, num? accessor(T))

Returns an unbiased estimator of the population variance of all values yielded by the accessor function applied to each element in the iterable using Welford’s algorithm.

This function ignores elements that yield values that do not satisfy any of the following conditions:

1. The value is not null.
2. The value is not double.nan.

Useful for filtering and ignoring missing data in datasets.

If the iterable has fewer than two elements that yield valid values, this function returns null.

Implementation

num? varianceBy<T>(Iterable<T> iterable, num? Function(T) accessor) {
  var count = 0;
  num delta, mean = 0, sum = 0;
  for (var element in iterable) {
    var value = accessor(element);
    if (value != null && value == value) {
      delta = value - mean;
      mean += delta / ++count;
      sum += delta * (value - mean);
    }
  }
  return count > 1 ? sum / (count - 1) : null;
}