multiple time-dependent covariates in cox regression

[gherma1]

I would like to enter multiple time-dependent covariates in a cox regression model.
Any suggestions on how to label the different time-dependent covariates because I receive errors with T_COV_A and T_COV_B?
Thanks

TIME PROGRAM.
COMPUTE T_COV_A = T_ *A T_COV_B = T_ *B
COXREG   csurvival
  /STATUS=Status(1)
  /CONTRAST (A)=Indicator
  /CONTRAST (B)=Indicator(1)
  /METHOD=ENTER A B T_COV_A T_CO_B
  /PRINT=CI(95)
  /CRITERIA=PIN(.05) POUT(.10) ITERATE(20).

>Error # 4285 in column 19.  Text: T_
>Incorrect variable name: either the name is more than 64 characters, or it is
>not defined by a previous command.
>Execution of this command stops.

[Bruce Weaver]

I don't have time to go generate and test an example right now, but the first thing that caught my eye was the missing command terminator on your COMPUTE line, and the fact that you attempt to compute two variables with a single COMPUTE command.  I.e., I wondered if this might work better:

TIME PROGRAM.
COMPUTE T_COV_A = T_ *A.
COMPUTE T_COV_B = T_ *B.
COXREG   csurvival
  /STATUS=Status(1)
  /CONTRAST (A)=Indicator
  /CONTRAST (B)=Indicator(1)
  /METHOD=ENTER A B T_COV_A T_CO_B
  /PRINT=CI(95)
  /CRITERIA=PIN(.05) POUT(.10) ITERATE(20).

HTH.

gherma1 wrote

I would like to enter multiple time-dependent covariates in a cox regression model.
Any suggestions on how to label the different time-dependent covariates because I receive errors with T_COV_A and T_COV_B?
Thanks

TIME PROGRAM.
COMPUTE T_COV_A = T_ *A T_COV_B = T_ *B
COXREG   csurvival
  /STATUS=Status(1)
  /CONTRAST (A)=Indicator
  /CONTRAST (B)=Indicator(1)
  /METHOD=ENTER A B T_COV_A T_CO_B
  /PRINT=CI(95)
  /CRITERIA=PIN(.05) POUT(.10) ITERATE(20).

>Error # 4285 in column 19.  Text: T_
>Incorrect variable name: either the name is more than 64 characters, or it is
>not defined by a previous command.
>Execution of this command stops.

[Bruce Weaver]

It just occurred to me that people not using NABBLE might get the BOLD codes coming through as asterisks.  So here are those two COMPUTE commands again without any BOLD codes:

COMPUTE T_COV_A = T_ *A.
COMPUTE T_COV_B = T_ *B.

Bruce Weaver wrote

I don't have time to go generate and test an example right now, but the first thing that caught my eye was the missing command terminator on your COMPUTE line, and the fact that you attempt to compute two variables with a single COMPUTE command.  I.e., I wondered if this might work better:

TIME PROGRAM.
COMPUTE T_COV_A = T_ *A.
COMPUTE T_COV_B = T_ *B.
COXREG   csurvival
  /STATUS=Status(1)
  /CONTRAST (A)=Indicator
  /CONTRAST (B)=Indicator(1)
  /METHOD=ENTER A B T_COV_A T_CO_B
  /PRINT=CI(95)
  /CRITERIA=PIN(.05) POUT(.10) ITERATE(20).

HTH.

gherma1 wrote

I would like to enter multiple time-dependent covariates in a cox regression model.
Any suggestions on how to label the different time-dependent covariates because I receive errors with T_COV_A and T_COV_B?
Thanks

TIME PROGRAM.
COMPUTE T_COV_A = T_ *A T_COV_B = T_ *B
COXREG   csurvival
  /STATUS=Status(1)
  /CONTRAST (A)=Indicator
  /CONTRAST (B)=Indicator(1)
  /METHOD=ENTER A B T_COV_A T_CO_B
  /PRINT=CI(95)
  /CRITERIA=PIN(.05) POUT(.10) ITERATE(20).

>Error # 4285 in column 19.  Text: T_
>Incorrect variable name: either the name is more than 64 characters, or it is
>not defined by a previous command.
>Execution of this command stops.

[gherma1]

Thank you very much for your reply

It worked with:
COMPUTE T_COV1_ = T_ * A.
COMPUTE T_COV2_ = T_ * B.

[Art Kendall]

COMPUTE T_COV_A = T_ *A T_COV_B = T_ *B
Haven't used this command. However looking at the syntax documentation, it may be that the transformationS must follow the usual syntax for COMPUTE.
See whether multiple COMPUTEs work for you.

[Bruce Weaver]

Nabble is saying that the OP's posts have not been delivered to the mailing list.  I Nabble is telling the truth about that, some of you will have missed the response quoted below.

gherma1 wrote

Thank you very much for your reply

It worked with:
COMPUTE T_COV1_ = T_ * A.
COMPUTE T_COV2_ = T_ * B.