ibeta function

double ibeta(num x, num a, num b)

Incomplete beta function.

Implementation

double ibeta(num x, num a, num b) {
  if (x < 0 || 1 < x) {
    return double.nan;
  }
  // Factor in front of the continued fraction.
  final bt = x == 0 || x == 1
      ? 0.0
      : exp(gammaLn(a + b) -
          gammaLn(a) -
          gammaLn(b) +
          a * log(x) +
          b * log(1.0 - x));
  if (x < (a + 1.0) / (a + b + 2.0)) {
    // Use continued fraction directly.
    return bt * betacf_(x, a, b) / a;
  } else {
    // Use continued fraction after making the symmetry transformation.
    return 1.0 - bt * betacf_(1.0 - x, b, a) / b;
  }
}