看到有点帖子里说到  伪R方值不能很好检验
也有说到用anova（）检验的，但是我看了下，R中的例子，看不懂，求大神解释。
data(barro)
fit0 <- rq(y.net ~  lgdp2 + fse2 + gedy2 , data = barro)
fit1 <- rq(y.net ~  lgdp2 + fse2 + gedy2 + Iy2 + gcony2, data = barro)
fit2 <- rq(y.net ~  lgdp2 + fse2 + gedy2 + Iy2 + gcony2, data = barro,tau=.75)
fit3 <- rq(y.net ~  lgdp2 + fse2 + gedy2 + Iy2 + gcony2, data = barro,tau=.25)
anova(fit0,fit1)
Quantile Regression Analysis of Deviance Table


Model 1: y.net ~ lgdp2 + fse2 + gedy2 + Iy2 + gcony2
Model 2: y.net ~ lgdp2 + fse2 + gedy2
  Df Resid Df F value    Pr(>F)   
1  2      155  18.879 4.596e-08 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


anova(fit1,fit2,fit3)
Quantile Regression Analysis of Deviance Table


Model: y.net ~ lgdp2 + fse2 + gedy2 + Iy2 + gcony2
Joint Test of Equality of Slopes: tau in {  0.5 0.75 0.25  }


  Df Resid Df F value  Pr(>F)  
1 10      473  1.8039 0.05751 .
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Warning message:
In summary.rq(x, se = se, covariance = TRUE) : 1 non-positive fis



不知道怎么看这个结果