Parametric Accelerated Failure Time Models
A Parametric Accelerated Failure Time (AFT) Model analysis invokes the parametric survival models procedure with nonrecurrent life time data. Parametric survival models assume that survival time follow a known distribution, and this analysis fits accelerated failure time models with their model effects proportional with respect to survival time.
Obtaining a Parametric Accelerated Failure Time Models analysis
This feature requires SPSS® Statistics Standard Edition or the Advanced Statistics Option.
- From the menus choose:
- Select a source variable.
- Time
-
- Survival
- Single numeric variable denoting the duration of the survival time.
- Start/End
- Numeric variables denoting Start Time and End Time.
- Status
- Single optional string or numeric variable that determines one of the following status settings:
- Failure/Event
- Maps a record to a failure/event category. The default value for a string status variable is F.
- Right Censoring
- Maps a record to a right censoring category. The default value for a string status variable is R.
- Left Censoring
- Maps a record to a left censoring category. The default value for a string status variable is L.
- Interval Censoring
- Maps a record to an interval censoring category. For Start/End only. The default value for a string status variable is I.
- Unmapped Values Treatment
- Controls which category to map the unmapped records to. To delete the records which failed to be mapped, select Exclude them from analysis.
- Covariate(s)
- One or more optional numeric variables to be treated as covariates. Note that a variable cannot be specified by both Covariate(s) and Fixed Factor(s).
- Fixed Factor(s)
- One or more optional variables to be treated as factors. A variable cannot be specified by both Fixed Factor(s) and Covariate(s).
- Left Truncation
- Single optional numeric variable for left truncation for Survival only.
This procedure pastes SURVREG AFT command syntax.