Showing Predictor "Toy dataset 4"

This predictor was successfully trained!

Meta

Notes:

Synthetic dataset from Hao, with 20 uncensored patients and 3 features (2nd feature redundant, 3rd feature a constant, so only 1 useful feature)

Publicly Accessible: yes

Viewable by Administrators: yes

Actions:Edit Delete

Hide

Dataset Show

Options
Regularization: L2	Tuning Score: avg log-likelihood loss
Time Points: 5	Tune Parameters: true
Feature Selection: false	Include Sibling Features: false
Standardize: true	Run Cross-Validation: true
Objective Function: log-likelihood	Marginal Loss Type: weighted
Use pre-defined Fold: No

Results
Start Time: 2017-09-22 11:42:34 -0600	Used Time: less than a minute
Label Name: SURVIVAL	C Parameter: 0.001
PSSP model: mtlr_model_file2843.mtlr	Version: 0.0.3

Predictor Settings / Retrain Show

Cross-validation results

5-Fold Cross-Validation Statistics*
Measure	PSSP Predictor (median)	PSSP Predictor (mean)	K-M Predictor
Concordance Index	1.0 ± 0.0	1.0 ± 0.0	0.5 ± 0.0
Hinged L1 Loss	1.25 ± 0.38		5.2 ± 0.29
Uncensored L1 loss	1.25 ± 0.38	2.41 ± 1.14	5.2 ± 0.29
Marginal L1 Loss	1.25 ± 0.38	2.41 ± 1.14	5.2 ± 0.29
Marginal L2 Loss	1.39 ± 0.42		5.92 ± 0.36
Hinged L1 Log-Loss	0.13 ± 0.05	0.27 ± 0.12	0.51 ± 0.05
Uncensored L1 Log-Loss	0.13 ± 0.05	0.27 ± 0.12	0.51 ± 0.05
Marginal L1 Log-Loss	0.11 ± 0.04	0.25 ± 0.11	0.47 ± 0.05
Log-Likelihood Loss	0.57 ± 0.35
D-calibration χ² statistics	22.0 ± 4.9		16.0 ± 0.0
D-calibration χ² p-value	0.34 ± 0.26		0.66 ± 0.0

*Mean value ± standard deviation.

Examine Classification Accuracy