In the linear regression quantile model, the conditional quantile of the response,
Y, given
x is
QY|x(τ)≡x′β(τ). Though
QY|x(τ) must be monotonically increasing, the Koenker–Bassett regression quantile estimator,
![View the MathML source View the MathML source](http://www.sciencedirect.com/cache/MiamiImageURL/B6V1D-4RFJ4FY-2-57/0?wchp=dGLbVzz-zSkWA)
, is not monotonic outside a vanishingly small neighborhood of
![View the MathML source View the MathML source](http://www.sciencedirect.com/cache/MiamiImageURL/B6V1D-4RFJ4FY-2-2/0?wchp=dGLbVzz-zSkWA)
. Given a grid of mesh
δn, let
![View the MathML source View the MathML source](http://www.sciencedirect.com/cache/MiamiImageURL/B6V1D-4RFJ4FY-2-59/0?wchp=dGLbVzz-zSkWA)
be the linear interpolation of the values of
a200b347afacd5fa590a02ce73">![View the MathML source View the MathML source](http://www.sciencedirect.com/cache/MiamiImageURL/B6V1D-4RFJ4FY-2-1K/0?wchp=dGLbVzz-zSkWA)
along the grid. We show here that for a range of rates,
δn,
![View the MathML source View the MathML source](http://www.sciencedirect.com/cache/MiamiImageURL/B6V1D-4RFJ4FY-2-5C/0?wchp=dGLbVzz-zSkWA)
will be strictly monotonic (with probability tending to one) and will be asymptotically equivalent to
![View the MathML source View the MathML source](http://www.sciencedirect.com/cache/MiamiImageURL/B6V1D-4RFJ4FY-2-4W/0?wchp=dGLbVzz-zSkWA)
in the sense that
n1/2 times the difference tends to zero at a rate depending on
δn.