Skip to main content

Table 3 The results of goodness-of-fit statistics and tests for four intercept-only models

From: Influencing factors of multiple adverse outcomes among schizophrenia patients using count regression models: a cross-sectional study

Models

Log likelihood

AIC

The likelihood ratio test (p-Value)a

The Vuong test

(p-Value)b

Aggressiveness with police dispatch or violent crime

 Poisson

-4827.2

9656.4

4893.7(< 0.001)

-4.2(< 0.001)

 NB

-2380.3

4764.7

—

-4.1(< 0.001)

 ZIP

-2870.0

5744.1

1121.8(< 0.001)

—

 ZINB

-2309.1

4624.3

—

—

Aggressiveness without police dispatch

 Poisson

-18,726.1

37,454.2

22,152.0(< 0.001)

-8.6(< 0.001)

 NB

-7650.2

15,304.4

—

-0.002 (= 0.499)

 ZIP

-10,786.3

21,576.6

6272.2(< 0.001)

—

 ZINB

-7650.2

15,306.4

—

—

Self-harm or suicide attempts

 Poisson

-990.9

1983.7

513.4(< 0.001)

-1.6(= 0.054)

 NB

-734.2

1472.4

—

-1.7(= 0.046)

 ZIP

-762.7

1529.3

92.8(< 0.001)

—

 ZINB

-716.3

1438.6

—

—

  1. NB negative binomial, ZIP zero-inflated Poisson, ZINB zero-inflated negative binomial
  2. athe likelihood ratio test for overdispersion (Poisson vs. NB and ZIP vs. ZINB)
  3. bthe Vuong test for excess zeros (Poisson vs. ZIP and NB vs. ZINB)