The coefficients are estimated at 95% CI. Coefficients of variables karno, squamous, and large celltype are found to be significant from the table below.
CoxPH Example Output Table (Columns 1-5)
| predictor |
category |
coefficient |
exp_coef |
std_error |
| karno |
|
-0.032815 |
0.967717 |
0.005508 |
| diagtime |
|
8.1e-05 |
1.000081 |
0.009136 |
| age |
|
-0.008706 |
0.991331 |
0.0093 |
| trt |
standard |
0 |
1 |
0 |
| trt |
test |
0.294603 |
1.342593 |
0.20755 |
| celltype |
adeno |
0 |
1 |
0 |
| celltype |
large |
-0.794775 |
0.451683 |
0.302878 |
| celltype |
smallcell |
-0.334506 |
0.715692 |
0.275978 |
| celltype |
squamous |
-1.196066 |
0.302381 |
0.300917 |
| prior |
no |
0 |
1 |
0 |
| prior |
yes |
0.071594 |
1.074219 |
0.232305 |
| Iteration # |
|
5 |
|
|
| Convergence |
|
|
|
|
| Likelihood ratio test |
|
62.1039 |
|
|
| Wald test |
|
62.3673 |
|
|
| Score test |
|
66.7375 |
|
|
CoxPH Example Output Table (Columns 6-9)
| std_error |
z_score |
p_value |
significance |
| 0.005508 |
-5.95802 |
0 |
*** |
| 0.009136 |
0.008901 |
0.992898 |
|
| 0.0093 |
-0.93615 |
0.349196 |
|
| 0 |
|
|
|
| 0.20755 |
1.419433 |
0.155773 |
|
| 0 |
|
|
|
| 0.302878 |
-2.624078 |
0.008688 |
** |
| 0.275978 |
-1.212075 |
0.225483 |
|
| 0.300917 |
-3.974739 |
7e-05 |
*** |
| 0 |
|
|
|
| 0.232305 |
0.308187 |
0.75794 |
|
| |
|
|
yes |
| |
|
0 |
on 8 degree of freedom |
| |
|
0 |
on 8 degree of freedom |
| |
|
0 |
on 8 degree of freedom |
The coefficients are output in table lungcancer_coef that is later used for prediction. As celltype, trt and prior are categorical variables, one of their categories is considered a reference for the other categories, thus trt = standard, celltype = adeno, and prior = no don’t show default coefficient values in each column.
The query below returns the output shown in the following two tables:
SELECT * FROM lungcancer_coef ORDER BY 1;
CoxPH Example Output Table lungcancer_coef (Columns 1-5)
| id |
predictor |
category |
coefficient |
exp_coef |
| 1 |
karno |
|
-0.0328153261941663 |
0.967717255116806 |
| 2 |
diagtime |
|
8.13205087074416e-05 |
1.00008132381531 |
| 3 |
age |
|
-0.00870647494549903 |
0.991331316650765 |
| 4 |
trt |
standard |
0 |
1 |
| 5 |
trt |
test |
0.294602821498042 |
1.34259300369677 |
| 6 |
celltype |
adeno |
0 |
1 |
| 7 |
celltype |
large |
-0.794774719851903 |
0.451682978670067 |
| 8 |
celltype |
smallcell |
-0.334505911425932 |
0.71569161405743 |
| 9 |
celltype |
squamous |
-1.19606637417932 |
0.302381330550752 |
| 10 |
prior |
no |
0 |
1 |
| 11 |
prior |
yes |
0.0715936019179389 |
1.07421869492581 |
CoxPH Example Output Table lungcancer_coef (Columns 6-9)
| std_error |
z_score |
p_value |
significance |
| 0.00550775688646227 |
-5.95802009250341 |
2.55312138097707e-09 |
*** |
| 0.00913606224777197 |
0.00890104582280767 |
0.992898086742234 |
|
| 0.00930029912031493 |
-0.936149991829963 |
0.349195966726043 |
|
| 0 |
NaN |
NaN |
|
| 0.207549603603519 |
1.41943331320844 |
0.155772725980423 |
|
| 0 |
NaN |
NaN |
|
| 0.30287771543449 |
-2.62407790124726 |
0.00868839104930008 |
** |
| 0.275977786191144 |
-1.21207549362053 |
0.225483483597614 |
|
| 0.300916994493076 |
-3.974738536101 |
7.04566159376308e-05 |
*** |
| 0 |
NaN |
NaN |
|
| 0.232305384067305 |
0.308187441308705 |
0.757939708088376 |
|
The linear predictor table lungcancer_lp is shown below.
SELECT * FROM lungcancer_lp ORDER BY 1;
CoxPH Example Output Table lungcancer_lp
| linear_predictor |
event |
time_interval |
| -4.41189466565077 |
1 |
467 |
| -4.41097447125404 |
1 |
110 |
| -4.39448171575977 |
1 |
389 |
| -4.29871049135928 |
1 |
283 |
| -4.28779288293039 |
0 |
182 |
| -4.26989200998715 |
1 |
143 |
| -4.25242310919077 |
1 |
999 |
| -4.20170368038227 |
0 |
25 |
| -4.10210453090365 |
0 |
100 |
| -4.04859022189228 |
1 |
112 |
| ... |
... |
... |