How do I formulate quantile regression as a Linear Programming problem? When looking at the median quantile problem I know it is
\begin{align} minimize \sum_{i=1}^{n}|\beta_{0} + X_{i}\beta_{1}-Y_{i}|\\ transforms\ into\\ minimize \sum_{i=1}^{n} e_{i}\\ s.t\\ e_{i}\geq \beta_{0} + X_{i}\beta_{1}-Y_{i}\\ e_{i}\geq -(\beta_{0} + X_{i}\beta_{1}-Y_{i}) \end{align}\begin{align} \text{minimize } & \sum_{i=1}^n |\beta_0 + X_i \beta_1-Y_i|\\ \text{transforms into } & \\ \text{minimize } & \sum_{i=1}^n e_i\\ \text{s.t.} & \\ & e_i\geq \beta_0 + X_i\beta_{1}-Y_i\\ & e_i\geq -(\beta_0 + X_i\beta_{1}-Y_i) \end{align} but how do I transform the minimization for other quantiles?