The quantile function is the inverse of the ogive, that is a linear
interpolation of the empirical quantile function.
The equation of the quantile function is
$$x = \frac{c_j (F_n(c_{j - 1}) - q) +
c_{j - 1} (q - F_n(c_j)}{F_n(c_j) - F_n(c_{j - 1})}$$
for \(0 \leq q \leq c_j\) and where \(c_0, \dots,
c_r\) are the \(r + 1\) group
boundaries and \(F_n\) is the empirical distribution function of
the sample.