Page 92 - Contributed Paper Session (CPS) - Volume 5
P. 92
CPS1108 Collins O. et al.
̂
( ) = + ∑ ∅(Λ − ) + ∑ ∅(Λ − ) + γ
t
0
=1 =1
Where is a link function and is a transformation
T
function. The parameter vector η = (η1, η2, …, ηr) corresponds to the effects of
covariates. To allow for regression on arbitrary past observations of the
response, define a set P = {i1, i2, …, ip}and integers 0 < i1 < i2 < … , with Pϵ .
This formulation is useful particularly when dealing with modeling stochastic
seasonality. Estimation and inference is derived by quasi conditional
maximum likelihood.
3. Result
Ante-Partum Haemorrhage
An antepartum haemorrhage (APH) refers to bleeding from the vagina that
occurs after the 20th week of pregnancy and before child-birth. The common
causes of bleeding during pregnancy are cervical ectropion, vaginal infection,
placental edge bleed, placenta praevia or placental abruption Cervix.
Count GLM for APH with SE and CI (level = 95 %) obtained by normal
approximation. Link function is log and distribution family is negative binomial.
The Log-likelihood value is -255. Score test on intervention(s) of given type at
given time is has p-value of < 0. Over dispersion coefficient σ was estimated to
2
be 0.018
Coefficient Estimate Std. Error CI(Lower) CI(Upper)
(Intercept) 3.033 0.6676 1.724 4.3412
Β1 0.367 0.139 0.095 0.6398
Intervention1 -0.189 0.0555 -0.298 -0.0805
(BeMOnC)
Intervention2 0.162 0.1653 -0.162 0.4858
(CeMOnC)
σ 0.017 N/A N/A N/A
2
Eclamsia
Refers to severe complication where high blood pressure results in seizures
during pregnancy. Seizures are periods of disturbed brain activity that can
cause episodes of staring, decreased alertness, and convulsions.
Count GLM for Eclamsia with SE and CI (level = 95 %) obtained by normal
approximation. Link function is log and distribution family is negative binomial.
The Log-likelihood value is -291.
Overdispersion coefficient σ was estimated to be 0.0622
2
Coefficient Estimate Std. Error CI(Lower) CI(Upper)
(Intercept) 3.6577 0.5855 2.51 4.8053
81 | I S I W S C 2 0 1 9