0
$\begingroup$

I'm trying to interpret these results of using R confint function , but I can not understand. This is a logistic regression about breast cancer. how to interpret confint function. What the 2.5% and 97.5% means?

logit1 = glm(goodmodel, data=train, family=binomial(link = logit))
summary(logit1)
confint(logit1)

 summary(logit1)

Call:
glm(formula = goodmodel, family = binomial(link = logit), data = train)

Deviance Residuals: 
     Min        1Q    Median        3Q       Max  
-1.76595  -0.00368  -0.00011   0.00000   2.57451  

Coefficients:
                          Estimate Std. Error z value Pr(>|z|)  
(Intercept)             -58.106512  37.736869  -1.540   0.1236  
radius_mean              -4.791077   1.933346  -2.478   0.0132 *
texture_mean              0.543094   0.222894   2.437   0.0148 *
compactness_mean        -93.504749  52.018859  -1.798   0.0723 .
`concave points_mean`   212.367971  86.957163   2.442   0.0146 *
radius_se                18.268981   7.771317   2.351   0.0187 *
smoothness_se           394.202770 347.004692   1.136   0.2560  
concavity_se            -78.413761  64.179173  -1.222   0.2218  
`concave points_se`     -26.914245 469.938101  -0.057   0.9543  
radius_worst              5.978607   4.897600   1.221   0.2222  
area_worst               -0.003889   0.045800  -0.085   0.9323  
concavity_worst          27.647216  14.104588   1.960   0.0500 *
symmetry_worst           25.702777  15.984453   1.608   0.1078  
fractal_dimension_worst  10.690937  79.071866   0.135   0.8924  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 533.015  on 397  degrees of freedom
Residual deviance:  26.608  on 384  degrees of freedom
AIC: 54.608

Number of Fisher Scoring iterations: 12

> confint(logit1)
Waiting for profiling to be done...
                                2.5 %        97.5 %
(Intercept)             -1.482910e+02    7.95727142
radius_mean             -9.791086e+00   -1.66857433
texture_mean             1.961861e-01    1.13920334
compactness_mean        -2.299043e+02   -5.16464149
`concave points_mean`    6.027894e+01  432.99676482
radius_se                6.704669e+00   39.25468272
smoothness_se           -1.961348e+02 1307.22045246
concavity_se            -3.122203e+02   25.69408799
`concave points_se`     -1.038278e+03  891.07772306
radius_worst            -2.977840e+00   17.32103028
area_worst              -9.549641e-02    0.09462597
concavity_worst          6.661555e+00   64.03373503
symmetry_worst           1.562622e+00   63.43243116
fractal_dimension_worst -1.619176e+02  168.07512057
There were 50 or more warnings (use warnings() to see the first 50)
````
$\endgroup$
2
  • 3
    $\begingroup$ Do you know what a confidence interval is? $\endgroup$ Jul 16, 2019 at 16:22
  • $\begingroup$ I'm still struggling with that, that's why I'm asking how to interpret this table, after that I'll probably have a clue. $\endgroup$
    – esteves13
    Jul 16, 2019 at 16:53

0

Browse other questions tagged or ask your own question.