Skip to main content

Table 1 Statistics for choosing the best number of classes

From: Influence of the trajectory of the urine output for 24 h on the occurrence of AKI in patients with sepsis in intensive care unit

Number of classes

Log likelihood

AIC

BIC

SABIC

Entropy

%class1

%class2

%class3

%class4

%class5

%class6

1

− 56976.9

113961.8

113990.6

113977.9

1.0000000

100.000000

     

2

− 53114.0

106244.1

106301.7

106276.2

0.9311908

8.997872

91.002128

    

3

− 51704.9

103433.7

103520.1

103482.0

0.9430789

88.570271

4.731989

6.697740

   

4

− 50503.1

101038.1

101153.3

101102.4

0.9150838

7.427298

82.095450

1.459114

9.018138

  

5

− 49930.4

99900.8

100044.8

99981.2

0.9182471

7.285439

80.565407

7.123315

3.556591

1.469247

 

6

− 49987.6

100023.2

100196.0

100119.7

0.8380689

9.038403

1.499645

0.000000

78.873239

9.585571

1.003141

  1. AIC: Akaike information criterion; BIC: Bayesian information criteria; SABIC: sample-adjusted information criteria