Quantile Regression
Fitting a line that a chosen fraction of outcomes falls below, using an asymmetric pinball loss instead of squared error, to predict a spread rather than only an average.
At quantile tau, underestimating the true value is penalized tau times as heavily as overestimating, which pushes the fitted line up or down until exactly that fraction of points falls at or below it. Fitting several quantiles independently captures heteroscedasticity — spread that changes with the input — directly from the data, without assuming any fixed distribution shape.
Independently-fit quantile lines can cross with limited data, since nothing in the basic method enforces that a higher quantile stays above a lower one.