Skip to main content

Table 2 Goodness of fit and predictive ability criterion of the tested ML-mixed models

From: Use of Maximum Likelihood-Mixed Models to select stable reference genes: a case of heat stress response in sheep

MID

n° parameters

-2LogL

AIC

BIC

BIC(%)

PD

PD(%)

1

5

-3141.0

-3097.0

-3081.4

87.67

0.001233

52.67

2

20

-3460.4

-3386.4

-3360.2

95.60

0.001230

52.54

3

36

-3658.2

-3552.2

-3514.7

100

0.001408

60.14

4

19

-3156.9

-3084.9

-3059.5

87.04

0.002027

86.58

5

35

-3203.0

-3099.0

-3062.2

87.12

0.002020

86.28

6

19

-3016.5

-2944.5

-2919.1

83.05

0.001827

78.04

7

35

-3308.0

-3204.0

-3167.2

90.11

0.002341

100

8

32

-3476.8

-3296.8

-3164.9

90.04

0.002084

89.02

  1. MID = model identification, -2LogL = -2 log of the likelihood function (smaller is better), AIC = Akaike Information Criterion (smaller is better), BIC = Bayes Information Criterion (smaller is better), BIC(%) = percentage of fit (higher is better), PD = predictive ability criterion (smaller is better), PD(%) = percentage of predictive ability loss (smaller is better)