Showing Predictor "MTLSAPaper_NSBCD"

This predictor was successfully trained!

Meta

Notes:

Test to see if we can run NSBCD dataset with 115 instances and 549 features (MTLSA paper claims that PSSP couldn't run high-dimensional data)

Publicly Accessible: yes

Viewable by Administrators: yes

Actions:Edit Delete

Hide

Dataset Show

Options
Regularization: L2	Tuning Score: avg log-likelihood loss
Time Points: 11	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: 2016-10-20 15:18:25 -0600	Used Time: 1 minute
Label Name: time	C Parameter: 100.0
PSSP model: mtlr_model_file2365.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	0.75 ± 0.07	0.75 ± 0.07	0.5 ± 0.0
Hinged L1 Loss	13.6 ± 2.72		15.5 ± 1.42
Uncensored L1 loss	33.41 ± 9.32	58.66 ± 6.7	34.67 ± 4.39
Marginal L1 Loss	33.35 ± 5.15	51.03 ± 7.32	22.01 ± 2.3
Marginal L2 Loss	N / A
Hinged L1 Log-Loss	0.39 ± 0.07	0.5 ± 0.06	0.44 ± 0.03
Uncensored L1 Log-Loss	1.03 ± 0.2	1.49 ± 0.12	1.17 ± 0.12
Marginal L1 Log-Loss	0.71 ± 0.12	0.94 ± 0.11	0.6 ± 0.05
Log-Likelihood Loss	1.54 ± 0.24
D-calibration χ² statistics	6.43 ± 2.53		9.61 ± 5.93
D-calibration χ² p-value	0.99 ± 0.02		0.86 ± 0.25

*Mean value ± standard deviation.

Examine Classification Accuracy