fixedLassoInf (error in output)

[Hayden-L-Smith]

When using "selectiveInference" after fitting a binary outcome in a LASSO model, I have noticed an error in the output when printing the generated object for the fixedLassoInf function. In particular, for protective covariates with both the "conf.low" and "conf.high" values below "1", the confidence interval values are missing the "-" sign. I looked in the code and was unable to see where this is happening at. For an example see below output for var 5 and 12.

 Var   Coef Z-score P-value LowConfPt UpConfPt LowTailArea UpTailArea
   1  0.408   5.033   0.002     0.211    0.542       0.049      0.048
   2  0.588   6.454   0.000     0.375    0.737       0.049      0.050
   3  0.085   1.198   0.215    -0.107    0.286       0.050      0.049
   4  0.345   4.783   0.014     0.102    0.461       0.049      0.048
   5 -0.444  -5.662   0.000     **0.314    0.573**       0.049      0.049
   6  0.094   1.377   0.169    -0.075    0.207       0.049      0.050
   7  0.346   4.944   0.000     0.231    0.463       0.050      0.049
   8  0.031   0.431   0.673    -0.486    0.131       0.050      0.048
   9  0.014   0.216   0.924    -2.283    0.036       0.050      0.049
  10  0.135   1.890   0.056    -0.007    0.346       0.049      0.050
  11  0.249   3.477   0.071    -0.036    0.362       0.049      0.049
  12 -0.174  -2.389   0.017     **0.042    0.301**       0.049      0.049

[matherealize]

I'd like to +1 the mentioned problem.

R 3.4.1
selectiveInference 1.2.4
glmnet 2.0-5

Binomial outcome variable. Workflow basically copied from the manual page of fixedLassoInf.

Example output (note. e.g. Variable 46 for an obvious example)

Testing results at lambda = 48.390, with alpha = 0.100

 Var   Coef Z-score P-value LowConfPt UpConfPt LowTailArea UpTailArea
   2  0.177   2.505   0.063    -0.017    0.332       0.049      0.049
   6 -0.163  -2.409   0.342    -0.323    0.265       0.050      0.050
  11 -0.111  -1.247   0.236    -0.163    0.373       0.050      0.050
  18 -0.175  -3.009   0.014     0.049    0.274       0.049      0.048
  21 -0.172  -2.714   0.685    -1.319    0.214       0.050      0.050
  22 -0.174  -2.812   0.079    -0.037    0.464       0.049      0.050
  24  0.259   3.595   0.266    -0.338    0.360       0.000      0.049
  26  0.095   1.592   0.125    -0.048    0.248       0.049      0.050
  29 -0.160  -2.613   0.321    -0.286    0.271       0.050      0.050
  33  0.222   2.927   0.016     0.063    0.535       0.049      0.049
  36  0.107   1.345   0.465    -0.403    0.225       0.050      0.049
  42  0.185   2.855   0.077    -0.034    0.301       0.049      0.049
  44 -0.229  -3.156   0.024     0.045    0.348       0.049      0.048
  45 -0.092  -1.030   0.254    -0.145    0.284       0.050      0.049
  46 -0.396  -6.804   0.000     0.300    0.492       0.049      0.049

Note: coefficients shown are partial regression coefficients