1.0 - 8.00 - CoxPH Output - Teradata Vantage

Teradata® Vantage Machine Learning Engine Analytic Function Reference

Teradata Vantage
Release Number
Release Date
May 2019
Content Type
Programming Reference
Publication ID
English (United States)

Summary Table Schema

Column Contents
predictor One row for each feature_column.
category Category name if predictor is categorical.
coefficient Estimated coefficient of input parameters.
exp_coef Exponent of coefficient (represents increase in units of hazard when covariate increases by 1).
std_error Standard error of coefficient.
z_score Wald test statistic for coefficient. (Wald test statistic assumes coefficient has normal distribution; that is, N(0, std_error).)
p_value p value for z_score. (p value represents significance of each individual coefficient.)
significance Significance code for p_value (see following table).

Significance Codes

p_value Significance Code
[0, 0.001) ***
[0.001, 0.01) **
[0.01, 0.05) *
[0.05, 0.1) .
[0.1, 1]  

Following the summary table, the function displays the values of the following:

  • Iteration#
  • Convergence
  • Likelikhood ratio test
  • Wald test
  • Score test
  • Degree of freedom

CoefficientTable Schema

Column Data Type Description
id INTEGER Row identifier
predictor VARCHAR Feature column name
coefficient DOUBLE PRECISION Estimated coefficient of the input parameters
exp_coef DOUBLE PRECISION Exponent of the coefficient
std_error DOUBLE PRECISION Standard error of the coefficient
z_score DOUBLE PRECISION Wald test statistic for the coefficient
p_value DOUBLE PRECISION p value for the z score
significance VARCHAR Significance code

LinearPredictorTable Schema

Column Data Type Description
linear_predictor DOUBLE PRECISION k fixed covariates and coefficients of input parameters.
event INTEGER 1 if event occurred during time interval, 0 otherwise.
time_interval INTEGER Time interval of input data; that is, end_time - start_time, in any time unit (for example, years, months, or days).
accumulate_column Same as in input table [Column appears once for each specified accumulate_column.] Column copied from input table.