r语言做非线性分位数回归出现 Error in rq.fit.br(x, y, tau = tau, ...) : Singular design matrix 此外: Warning messages:1: In lsfit(x, y, d^2, intercept = FALSE) : 'X' matrix was collinear
2: In lsfit(x, y, d^2, intercept = FALSE) : 'X' matrix was collinear
请问该如何改进？这个错误有时候出现，有时候不出现。若能解答，我会给奖励