r - Unexpected quantile relationship -

i want fit quantile regression model observed data, show triangular relationship between response , predictor variables:

when do:

library("quantreg") m1 <- rq(y~ x, tau = 0.75, data=mydata)  summary(m1)  call: rq(formula = y ~ x, tau = 0.75, data = mydata)  tau: [1] 0.75  coefficients:              coefficients lower bd upper bd (intercept) 3.42758      1.80850  4.74463   x           0.27879      0.07132  0.82591  

it founds positive relationship (in red), when should negative looking @ points in graph, right? maybe i'm missing looks wrongh tau value. tried t=0.97 , t=0.90 (in gray), same pattern produced.

then, when do:

m1.all <- rq(y~ x, tau = seq(0.05, 0.95, = 0.05), data=mydata)  m1.plot <- summary(m1.all) 

warning messages:
1: in rq.fit.br(x, y, tau = tau, ci = true, ...) :
solution may nonunique
2: in rq.fit.br(x, y, tau = tau, ci = true, ...) :
solution may nonunique
3: in rq.fit.br(x, y, tau = tau, ci = true, ...) :
solution may nonunique
4: in rq.fit.br(x, y, tau = tau, ci = true, ...) :
solution may nonunique

plot(m1.plot) 

error in plot.window(...) : infinite axis extents [gepretty(-inf,inf,5)]

i obtain plot intercept, not coefficients.

what i'm doing wrong?

i provide here mydata. expect negative relationship similar results shown cade & noon 2003 in fig. 1 (see here).