Showing Predictor "NACD L2-reg L2-loss(w), noFS"

This predictor was successfully trained!

Meta

Notes:

Based on 2104 .. but using L2 marginal loss

(I thought 2104 used that loss, but I guess not.)

Publicly Accessible: yes

Viewable by Administrators: yes

Parent Predictor: Show

Actions:Edit Delete

Hide

Dataset Show

Options
Regularization: L2	Tuning Score: avg log-likelihood loss
Time Points: 50	Tune Parameters: true
Feature Selection: false	Include Sibling Features: false
Standardize: false	Run Cross-Validation: true
Objective Function: log-likelihood	Marginal Loss Type: weighted
Use pre-defined Fold: No

Results
Start Time: 2016-06-30 12:55:49 -0600	Used Time: 32 minutes
Label Name: SURVIVAL	C Parameter: 0.1
PSSP model: mtlr_model_file2105.mtlr	Version: 0.0.1

Predictor Settings / Retrain Show

Cross-validation results

5-Fold Cross-Validation Statistics*
Measure	PSSP Predictor (median)	PSSP Predictor (mean)	K-M Predictor
Concordance Index	0.76 ± 0.0		0.5 ± 0.0
Hinged L1 Loss	9.77 ± 0.23		13.28 ± 0.05
Uncensored L1 loss	11.3 ± 0.26		12.18 ± 0.05
Marginal L1 Loss	15.63 ± 0.28		20.72 ± 0.14
Marginal L2 Loss	N / A
Hinged L1 Log-Loss	0.56 ± 0.02		0.79 ± 0.01
Uncensored L1 Log-Loss	0.76 ± 0.02		0.97 ± 0.01
Marginal L1 Log-Loss	0.73 ± 0.02		1.01 ± 0.01
Log-Likelihood Loss	2.62 ± 0.02
D-calibration χ² statistics	15.07 ± 2.29		0.97 ± 0.12
D-calibration χ² p-value	0.7 ± 0.13		1.0 ± 0.0

*Mean value ± standard deviation.

Examine Classification Accuracy