Skip to main content

Table 4 Comparison of hidden Markov models fitted to parasite and leukocyte counts by AIC and BIC

From: Evidence for overdispersion in the distribution of malaria parasites and leukocytes in thick blood smears

 

Poisson HMM

Negative binomial HMM

m =1

-ℒ

AIC

BIC

-ℒ

AIC

BIC

p a

6801.59

13605.17

13609.80

3200.63

6405.25

6414.50

p b

10838.95

21679.91

21684.75

4344.27

8692.54

8702.23

p c

2472.18

4946.36

4951.08

2302.96

4609.92

4619.38

ℓ a

3108.25

6218.51

6223.13

2532.77

5069.53

5078.79

ℓ b

3547.53

7097.06

7101.90

2965.34

5934.69

5944.38

ℓ c

3051.08

6104.15

6108.88

2728.46

5460.91

5470.37

m =2

-ℒ

AIC

BIC

-ℒ

AIC

BIC

p a

3877.14

7764.27

7787.40

3043.31

6098.62

6126.37

p b

5794.89

11599.77

11623.99

4166.23

8344.45

8373.51

p c

2228.73

4467.47

4491.11

2224.71

4461.42

4489.79

ℓ a

2578.83

5167.66

5190.79

2433.86

4879.72

4907.47

ℓ b

2993.67

5997.35

6021.57

2889.88

5791.76

5820.82

ℓ c

2667.70

5345.41

5369.05

2640.61

5293.22

5321.59

m =3

-ℒ

AIC

BIC

-ℒ

AIC

BIC

p a

6447.60

3265.54

6553.09

6603.97

3008.87

6035.74

p b

4634.75

9291.50

9344.78

4126.32

8270.64

8314.23

p c

2210.74

4443.48

4495.49

2215.95

4449.90

4492.46

ℓ a

2414.70

4851.41

4902.28

2394.82

4807.64

4849.27

ℓ b

2898.08

5818.17

5871.45

2884.03

5786.06

5829.65

ℓ c

2609.50

5241.00

5293.01

2619.57

5257.14

5299.69

m =4

-ℒ

AIC

BIC

-ℒ

AIC

BIC

p a

3096.91

6231.82

6319.70

2985.36

5994.73

6050.23

p b

4322.77

8683.53

8775.57

4117.57

8259.14

8317.27

p c

2206.93

4451.87

4541.71

2214.22

4452.45

4509.19

ℓ a

2380.19

4798.38

4886.26

2390.87

4805.74

4861.24

ℓ b

2880.72

5799.44

5891.48

2881.97

5787.95

5846.07

ℓ c

2599.52

5237.05

5326.89

2615.98

5255.96

5312.71

  1. Parasite (p a , p b , p c ) and leukocyte ( ℓ a , ℓ b , ℓ c ) counts are fitted to Poisson HMMs and negative binomial HMMs. The number of components is m . Minus log-likelihood (-ℒ) and information measures (AIC and BIC) are given. Models were fitted by maximum likelihood using the expectation-maximization (EM) algorithm, and validated by direct numerical maximization using nlm in R.
Back to article page

Malaria Journal

ISSN: 1475-2875

Contact us