Fabien COELHO and Robert Haas
(PHI(2.0 * threshold * (i - min - mu + 0.5) / (max - min + 1)) -
PHI(2.0 * threshold * (i - min - mu - 0.5) / (max - min + 1))) /
- (2.0 * PHI(threshold) - 1.0)
- >
+ (2.0 * PHI(threshold) - 1.0)>.
Intuitively, the larger the threshold>, the more
frequently values close to the middle of the interval are drawn, and the
less frequently values close to the min> and