Test Data for rpasurvival Function

Synthetic datasets for testing and demonstrating the rpasurvival function (Recursive Partitioning Analysis for Survival Data).

Format

rpasurvival_test: Standard dataset with 200 observations and 11 variables:

patient_id: Character. Patient identifier (PT001-PT200)
time: Numeric. Survival time in months (range: 0.5-120, mean ~36)
event: Factor. Event indicator (0 = censored, 1 = death/event). Event rate ~65%
age: Numeric. Patient age in years (40-85, mean ~65)
stage: Ordered factor. Tumor stage (I, II, III, IV)
grade: Ordered factor. Tumor grade (G1, G2, G3)
LVI: Factor. Lymphovascular invasion (Absent, Present)
tumor_size: Numeric. Tumor size in centimeters (0.5-10)
ki67: Numeric. Ki-67 proliferation index, percentage (0-100). ~3% missing
performance_status: Ordered factor. ECOG performance status (0, 1, 2)
treatment: Factor. Treatment modality (Surgery only, Surgery + Chemo, Surgery + Radio, Trimodal)

rpasurvival_small: Minimal dataset with 50 observations and 6 variables:

patient_id: Character. Patient identifier (SM01-SM50)
time: Numeric. Survival time in months
event: Factor. Event indicator (0, 1)
age: Numeric. Patient age in years
stage: Factor. Tumor stage (Early, Advanced)
grade: Factor. Tumor grade (Low, High)

rpasurvival_large: Large dataset with 500 observations and 11 variables:

patient_id: Character. Patient identifier (LG0001-LG0500)
time: Numeric. Survival time in months
event: Factor. Event indicator (0, 1). Event rate ~70%
age: Numeric. Patient age in years
stage: Ordered factor. Detailed tumor stage (IA, IB, IIA, IIB, IIIA, IIIB, IV)
grade: Ordered factor. Tumor grade (1, 2, 3)
LVI: Factor. Lymphovascular invasion (No, Yes)
PNI: Factor. Perineural invasion (No, Yes)
tumor_size: Numeric. Tumor size in centimeters
nodes_positive: Numeric. Number of positive lymph nodes
biomarker1: Numeric. Continuous biomarker 1
biomarker2: Numeric. Continuous biomarker 2

Edge case datasets (for testing different event/time coding):

rpasurvival_edge_truefalse: 30 observations with event coded as TRUE/FALSE
rpasurvival_edge_12: 30 observations with event coded as 1/2
rpasurvival_edge_days: 30 observations with time in days
rpasurvival_edge_years: 30 observations with time in years

Source

Generated synthetically using data-raw/rpasurvival_test_data.R. Seed: 12345. Generation date: 2026-01-31.

Details

These datasets were generated using a seeded random number generator to produce realistic survival data with the following characteristics:

Survival times follow exponential distribution
Event rates are clinically realistic (60-70%)
Prognostic correlations built in (Stage IV → shorter survival)
Missing data pattern (~3% in continuous biomarkers)
Events-per-variable (EPV) ratio > 10 for all datasets

The data generation process ensures:

Non-negative survival times
Proper factor level ordering (ordinal variables)
Realistic clinical distributions
Sufficient sample sizes for RPA analysis

File Formats

Each dataset is available in multiple formats:

RDA: Native R format (use data())
CSV: Comma-separated values
XLSX: Excel format
OMV: jamovi native format

Usage Examples

See vignette("rpasurvival-examples") for comprehensive examples.

Basic usage:


data(rpasurvival_test)
library(ClinicoPath)

# Standard RPA analysis
rpasurvival(
  data = rpasurvival_test,
  time = "time",
  event = "event",
  predictors = c("age", "stage", "grade", "LVI"),
  time_unit = "months"
)

# Test small sample warnings
data(rpasurvival_small)
rpasurvival(
  data = rpasurvival_small,
  time = "time",
  event = "event",
  predictors = c("stage", "grade")
)

# Test different event coding
data(rpasurvival_edge_truefalse)
rpasurvival(
  data = rpasurvival_edge_truefalse,
  time = "time",
  event = "event_tf",
  predictors = c("stage", "grade"),
  eventValue = "TRUE"
)

Testing Scenarios

The datasets support testing of:

Standard analysis: Use rpasurvival_test with 4-6 predictors
Small samples: Use rpasurvival_small, expect warnings
Complex trees: Use rpasurvival_large with maxdepth=5
Event coding: Test TRUE/FALSE and 1/2 coding schemes
Time units: Test days, months, years with time_unit parameter
Missing data: Verify handling of ~3% missing values
Mixed predictors: Continuous, ordinal, and nominal variables

Validation

All datasets have been validated for:

Non-negative survival times
Appropriate event rates
Stage-survival correlation (higher stage → worse prognosis)
Sufficient EPV (events per variable > 10)
Realistic clinical distributions
Proper factor level ordering

References

Liu Y, et al. (2026). Recursive partitioning analysis for survival data.

Examples

# Load standard test data
data(rpasurvival_test)

# Examine structure
str(rpasurvival_test)
#> tibble [200 × 11] (S3: tbl_df/tbl/data.frame)
#>  $ patient_id        : chr [1:200] "PT001" "PT002" "PT003" "PT004" ...
#>  $ time              : num [1:200] 9.5 11.4 17.5 0.3 16.4 ...
#>  $ event             : Factor w/ 2 levels "0","1": 1 2 1 2 1 1 1 2 2 1 ...
#>  $ age               : num [1:200] 81 62 59 66 66 69 74 63 80 57 ...
#>  $ stage             : Ord.factor w/ 4 levels "I"<"II"<"III"<..: 3 3 3 4 1 3 1 2 4 1 ...
#>  $ grade             : Ord.factor w/ 3 levels "G1"<"G2"<"G3": 2 1 1 3 2 2 3 3 2 2 ...
#>  $ LVI               : Factor w/ 2 levels "Absent","Present": 2 1 1 1 1 1 2 2 2 2 ...
#>  $ tumor_size        : num [1:200] 2.43 3.72 7.11 3.4 0.5 ...
#>  $ ki67              : num [1:200] 32 25.4 37.3 61.8 NA ...
#>  $ performance_status: Ord.factor w/ 3 levels "0"<"1"<"2": 3 3 1 2 2 1 2 2 3 1 ...
#>  $ treatment         : Factor w/ 4 levels "Surgery + Chemo",..: 1 3 3 1 4 1 1 3 2 4 ...

# Summary statistics
summary(rpasurvival_test)
#>   patient_id             time         event        age        stage    grade  
#>  Length:200         Min.   :  0.300   0: 70   Min.   :37.00   I  :49   G1:42  
#>  Class :character   1st Qu.:  7.775   1:130   1st Qu.:56.00   II :52   G2:92  
#>  Mode  :character   Median : 15.800           Median :63.00   III:74   G3:66  
#>                     Mean   : 26.020           Mean   :63.45   IV :25          
#>                     3rd Qu.: 36.200           3rd Qu.:72.00                   
#>                     Max.   :230.500           Max.   :90.00                   
#>                                                                               
#>       LVI        tumor_size         ki67       performance_status
#>  Absent :114   Min.   :0.500   Min.   : 0.00   0:97              
#>  Present: 86   1st Qu.:2.446   1st Qu.:15.75   1:63              
#>                Median :3.651   Median :30.42   2:40              
#>                Mean   :3.764   Mean   :30.80                     
#>                3rd Qu.:5.089   3rd Qu.:44.55                     
#>                Max.   :8.597   Max.   :97.00                     
#>                NA's   :6       NA's   :6                         
#>            treatment 
#>  Surgery + Chemo:77  
#>  Surgery + Radio:34  
#>  Surgery only   :62  
#>  Trimodal       :27  
#>                      
#>                      
#>                      

# Check event rate
table(rpasurvival_test$event)
#> 
#>   0   1 
#>  70 130 
prop.table(table(rpasurvival_test$event))
#> 
#>    0    1 
#> 0.35 0.65 

# Check stage distribution
table(rpasurvival_test$stage)
#> 
#>   I  II III  IV 
#>  49  52  74  25 

# Basic RPA analysis
if (FALSE) { # \dontrun{
library(ClinicoPath)
result <- rpasurvival(
  data = rpasurvival_test,
  time = "time",
  event = "event",
  predictors = c("age", "stage", "grade", "LVI")
)
} # }