Epidemic model
Communicable diseases such as measles, influenza, and tuberculosis are a fact of life. We
will be concerned with both epidemics, which are sudden outbreaks of a disease, and
endemic situations, in which a disease is always present. The AIDS epidemic, the recent
SARS epidemic, recurring influenza pandemics, and outbursts of diseases such as the Ebola
virus are events of concern and interest to many people. The prevalence and effects of
many diseases in less-developed countries are probably not as well known but may be of
even more importance. Every year millions, of people die of measles, respiratory infections,
diarrhoea, and other diseases that are easily treated and not considered dangerous in the
Western world. Diseases such as malaria, typhus, cholera, and sleeping sickness are
endemic in many parts of the world. The effects of high disease mortality on mean life span
and of disease debilitation and mortality on the economy in afflicted countries are
considerable.
An epidemic, which acts on a short temporal scale, may be described as a sudden outbreak
of a disease that infects a substantial portion of the population in a region before it
disappears. Epidemics usually leave many members untouched. Often these attacks recur
with intervals of several years between outbreaks, possibly diminishing in severity as
populations develop some immunity. This is an important aspect of the connection between
epidemics and disease evolution.
Mathematical modelling in epidemiology provides understanding of the underlying
mechanisms that influence the spread of disease, and in the process, it suggests control
strategies. In fact, models often identify behaviours that are unclear in experimental data–
often because data are no reproducible and the number of data points is limited and subject
to errors in measurement. For example, one of the fundamental results in mathematical
epidemiology is that most mathematical epidemic models, including those that include a
high degree of heterogeneity, usually exhibit “threshold” behaviour, which in
epidemiological terms can be stated as follows: If the average number of secondary
infections caused by an average infective is less than one, a disease will die out, while if it
exceeds one there will be an epidemic. This broad principle, consistent with observations
and quantified via epidemiological models, has been used routinely to estimate the
effectiveness of vaccination policies and the likelihood that a disease may be eliminated or
eradicated. Hence, even if it is not possible to verify hypotheses accurately, agreement with
hypotheses of a qualitative nature is often valuable. Expressions for the basic reproductive
number for HIV in various populations are being used to test the possible effectiveness of
vaccines that may provide temporary protection by reducing either HIV-infectiousness or
susceptibility to HIV. Models are used to estimate how widespread a vaccination plan must
be to prevent or reduce the spread of HIV.
Hence we can say that epidemic models are used to study the effect of infectious disease.
Page 1 of 15
The spread of infectious can occur in two ways
(1) Direct way/contact: occurs when there is physical contact between an infected
person and susceptible person.
Direct contact infections spread when disease-causing microorganisms pass from the
infected person to the healthy person via direct physical contact with blood or body
fluids.
Example: touching, kissing, sexual contact, contact with oral secretions or contact
with body lesions.
(1) In Direct way/contact: occurs when there is no direct human-to-human contact.
Contact occurs from a reservoir to contaminated surface or objects or to vectors
such as mosquitoes, flies, mites, fleas, ticks, rodents or dogs.
In Direct contact infections spread when an infected person sneezes or coughs,
sending infectious droplets into air. If healthy people inhale the infectious droplets,
or if the contaminated droplets land directly in their eyes, nose or mouth, they risk
becoming ill.
In order to model such an epidemic, we divide the population being studied into three
classes labelled
(i) Population of infective (cause)(S): infectious and able to spread the disease
by contact with susceptible.
(ii) Population of susceptible(are at risk)(I): those who are at Risk
(iii) Population of isolated/removed individuals (death/recovery)(R): isolated
from the group either due to recover or death
S, I, and R. We let X(t) denote the number of individuals who are susceptible to the disease,
that is, who are not yet infected at time t. Y(t) denotes the number of infected individuals,
assumed infectious and able to spread the disease by contact with susceptible. Z(t) denotes
the number of individuals who have been infected and then removed from the possibility of
being infected again or of spreading infection. Removal is carried out through isolation from
the rest of the population, through immunization against infection, through recovery from
the disease with full immunity against re-infection, or through death caused by the disease.
These characterizations of removed members are different from an epidemiological
perspective but are often equivalent from a modelling point of view that takes into account
only the state of an individual with respect to the disease.
Page 2 of 15
I S R
Simple Epidemic model:
Consider a closed population of n individuals.
indivi There are two states:
Susceptible
Infected
Initially 𝑌(𝑡) are infected n – 𝑌(𝑡𝑡) are therefore susceptible.
We assume the population is well mixed. The probability that a susceptible and infectious
individual meet is proportional to their abundances, with effective transmission rate β.
Let X(t) and Y(t) be the number of susceptible and number of infected indiv
individuals
iduals at time t
respectively. X(t) and Y(t) are positive integer.
Let β be the time independent infection rate i.e, rate at which susceptible get infected.
𝑃 (𝐴 𝑝𝑒𝑟𝑠𝑜𝑛 𝑔𝑒𝑡𝑡𝑖𝑛𝑔 𝑖𝑛𝑓𝑒𝑐
𝑖𝑛𝑓𝑒𝑐𝑡𝑒𝑑 𝑖𝑛 𝑡 𝑡𝑜 𝑡 + ∆𝑡) = 𝛽𝑋(𝑌(𝑡)𝑡)∆𝑡 + 𝑜(∆𝑡) (1)
Now let us assume that an epidemic starts with one infected and n susceptible i.e,
𝑋(0) = 𝑛 𝑎𝑛𝑑 𝑌(0) = 1
𝑋(0) + 𝑌(0) = 𝑛 + 1
So the total population be (n+1).
Probability of r susceptible
ble in time t will be given by,
𝑝 (𝑡)
= 𝑃(𝑟 𝑠𝑢𝑠𝑝𝑒𝑐𝑡𝑖𝑏𝑙𝑟𝑒 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡 𝑔𝑖𝑣𝑒𝑛 𝑡ℎ𝑎𝑡 𝑖𝑛𝑖𝑡𝑖𝑎𝑙𝑙𝑦 𝑡ℎ𝑒𝑟𝑒 𝑤𝑒𝑟𝑒 𝑛 𝑠𝑢𝑐𝑝𝑒𝑡𝑖𝑏𝑙𝑒 𝑎𝑛𝑑 1 𝑖𝑛𝑓𝑒𝑐𝑡𝑖𝑜𝑢𝑠)
( )
𝑝 (𝑡) = 𝑃 𝑋(𝑡) = 𝑟 𝑋(0) = 𝑛 𝑎𝑛𝑑 𝑌(0) = 1 (2)
Then
Page 3 of 15
𝑝 (𝑡 + ∆𝑡) = 𝑃(𝑟 𝑠𝑢𝑠𝑝𝑒𝑐𝑡𝑖𝑏𝑙𝑒 𝑎𝑡 𝑡𝑖𝑚𝑒 𝑡 𝑎𝑛𝑑 𝑛𝑜 𝑖𝑛𝑓𝑒𝑐𝑡𝑖𝑜𝑛 𝑖𝑛 (𝑡
+ ∆𝑡) 𝑜𝑟 𝑟 𝑠𝑢𝑠𝑝𝑒𝑐𝑡𝑖𝑏𝑙𝑒 𝑡𝑖𝑙𝑙 𝑡𝑖𝑚𝑒 𝑡 𝑎𝑛𝑑 1 𝑖𝑛𝑓𝑒𝑐𝑡𝑖𝑜𝑛 𝑖𝑛 𝑡 𝑡𝑜 (𝑡
+ ∆𝑡) 𝑔𝑖𝑣𝑒𝑛 𝑡ℎ𝑎𝑡 𝑖𝑛𝑖𝑡𝑖𝑎𝑙𝑙𝑦 𝑡ℎ𝑒𝑟𝑒 𝑤𝑒𝑟𝑒 𝑛 𝑠𝑢𝑐𝑝𝑒𝑡𝑖𝑏𝑙𝑒 𝑎𝑛𝑑 1 𝑖𝑛𝑓𝑒𝑐𝑡𝑖𝑜𝑢𝑠)
= 𝑝 (𝑡)[1 − 𝛽𝑌(𝑡 )𝑋(𝑡)∆𝑡 + 𝑜(∆𝑡)] + 𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡)∆𝑡 + 𝑜(∆𝑡)
= 𝑝 (𝑡) − 𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡)∆𝑡 + 𝑜(∆𝑡) + 𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡)∆𝑡 + 𝑜(∆𝑡)
𝑝 (𝑡 + ∆𝑡) − 𝑝 (𝑡) = −𝑝 (𝑡 )𝛽𝑌(𝑡 )𝑋(𝑡)∆𝑡 + 𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡)∆𝑡 + 𝑜(∆𝑡)
𝑝 (𝑡 + ∆𝑡) − 𝑝 (𝑡) = −𝑝 (𝑡 )𝛽𝑌(𝑡 )𝑋(𝑡)∆𝑡 + 𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡)∆𝑡 + 𝑜(∆𝑡)
𝑝 (𝑡 + ∆𝑡) − 𝑝 (𝑡)
= −𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡) + 𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡) + 𝑜(∆𝑡)
∆𝑡
If ∆𝑡 → 0
𝑝′ (𝑡) = −𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡) + 𝑝 (𝑡)𝛽𝑌(𝑡)𝑋(𝑡)
𝑝′ (𝑡) = −𝑝 (𝑡)𝛽𝑟(𝑛 − 𝑟 + 1) + 𝑝 (𝑡)𝛽(𝑟 + 1)(𝑛 − 𝑟) (3)
Putting 𝜏 = 𝛽𝑡
𝑝 (𝜏) = 𝛽𝑝 (𝑡)
𝛽𝑝 (𝑡 ) = 𝑝 (𝜏)
𝑑𝛽𝑝 (𝑡 ) 𝑑𝑝 (𝜏)
=
𝑑𝑡 𝑑𝑡
𝑑𝑝 (𝑡) 𝑑𝑝 (𝜏) 𝑑𝜏
𝛽 =
𝑑𝑡 𝑑𝜏 𝑑𝑡
𝑑𝑝 (𝑡 ) 𝑑𝑝 (𝜏)
= (4)
𝑑𝑡 𝑑𝜏
𝑑𝑝 (𝜏)
= −𝑝 (𝜏)𝛽𝑟(𝑛 − 𝑟 + 1) + 𝑝 (𝜏)𝛽(𝑟 + 1)(𝑛 − 𝑟) (5)
𝑑𝜏
0≤𝑟≤𝑛
Let us take r=n
𝑑𝑝 (𝜏)
= −𝑝 (𝜏)𝛽𝑟
𝑑𝜏
𝑝′ (𝜏) = −𝑝 (𝜏)𝛽𝑟
Laplace Transformation
Page 4 of 15
Laplace transform is the integral transform of the given derivative function with real
variable t to convert into complex function with variable s. For t ≥ 0, let f(t) be given and
assume the function satisfies certain conditions.
The Laplace transform is a well established mathematical technique for solving a differential
equation. Many mathematical problems are solved using transformations. The idea is to
transform the problem into another problem that is easier to solve.
The Laplace transform of f(t), that it is denoted by f(t) or F(s) is defined by the equation
𝐿{𝑓(𝑡)} = 𝑒 𝑓(𝑡) 𝑑𝑡
Example: if
f(t) = 1, 𝑡 ≥ 0
then Laplace transformation
𝐿{𝑓(𝑡)} = 𝑒 𝑑𝑡
𝑒 1
= − =
𝑠 𝑠
( )
𝐿{𝑓(𝑡)} = 𝑒 ∗𝑒 𝑑𝑡 = 𝑒 𝑑𝑡
( )
𝑒 1
= − =
𝑠+𝑎 𝑠+𝑎
Laplace transformation of derivative
𝐿{𝑓′(𝑡)} = 𝑒 𝑓′(𝑡) 𝑑𝑡
=𝑒 𝑓′(𝑡) 𝑑𝑡 − 𝑓′(𝑡)𝑒 𝑑𝑡
= [𝑒 𝑓(𝑡)] + 𝑠 𝑒 𝑓(𝑡) 𝑑𝑡
= [𝑒 𝑓(∞) − 𝑒 𝑓(0)] + 𝑠 𝑒 𝑓(𝑡) 𝑑𝑡
𝐿{𝑓′(𝑡)} = [0 − 𝑓 (0)] + 𝑠𝐿{𝑓(𝑡)}
Page 5 of 15
𝐿{𝑓′(𝑡)} = 𝑠𝐿{𝑓(𝑡)} − 𝑓(0)
𝑝′ (𝜏) = −𝑝 (𝜏)𝛽𝑟
Taking Laplace transformation
𝐿{𝑝′ (𝜏)} = 𝐿{−𝑝 (𝜏)𝛽𝑟}
𝑠𝐿{𝑝 (𝜏)} − 𝑝 (0) = −𝛽𝑟𝐿{𝑝 (𝜏)}
𝑠𝐿{𝑝 (𝜏)} + 𝛽𝑟𝐿{𝑝 (𝜏)} = 𝑝 (0)
𝑠𝐿{𝑝 (𝜏)} + 𝛽𝑟𝐿{𝑝 (𝜏)} = 1
[𝑠 + 𝛽𝑟]𝐿{𝑝 (𝜏)} = 1
1
𝐿{𝑝 (𝜏 )} =
𝑠 + 𝛽𝑟
Now using inverse Laplace transformation
1
𝑝 (𝜏) = 𝐿
𝑠 + 𝛽𝑟
𝑝 (𝑡) = 𝑒 (6)
Case II: when r<n
Consider (2)
𝜏 = 𝛽𝑡
𝑝′ (𝑡) = −𝑝 (𝑡)𝛽𝑟(𝑛 − 𝑟 + 1) + 𝑝 (𝑡)𝛽(𝑟 + 1)(𝑛 − 𝑟) (2)
𝑝′ (𝜏) = −𝑝 (𝜏)𝑟(𝑛 − 𝑟 + 1) + 𝑝 (𝜏)(𝑟 + 1)(𝑛 − 𝑟) (2)
Taking Laplace on both sides
𝐿{𝑝′ (𝜏)} = −𝑟(𝑛 − 𝑟 + 1)𝐿{𝑝 (𝜏)} + (𝑟 + 1)(𝑛 − 𝑟)𝐿{𝑝 (𝜏)}
Laplace transformation of derivative
𝐿{𝑓′(𝑡)} = 𝑒 𝑓′(𝑡) 𝑑𝑡
=𝑒 𝑓′(𝑡) 𝑑𝑡 − 𝑓′(𝑡)𝑒 𝑑𝑡
Page 6 of 15
= [𝑒 𝑓(𝑡)] + 𝑠 𝑒 𝑓(𝑡) 𝑑𝑡
= [𝑒 𝑓(∞) − 𝑒 𝑓(0)] + 𝑠 𝑒 𝑓(𝑡) 𝑑𝑡
𝐿{𝑓′(𝑡)} = [0 − 𝑓(0)] + 𝑠𝐿{𝑓(𝑡)}
𝐿{𝑓′(𝑡)} = 𝑠𝐿{𝑓(𝑡)} − 𝑓(0)
𝑠𝐿{𝑝 (𝜏)} − 𝑝 (0) = −𝑟(𝑛 − 𝑟 + 1)𝐿{𝑝 (𝜏)} + (𝑟 + 1)(𝑛 − 𝑟)𝐿{𝑝 (𝜏 )}
𝑠𝐿{𝑝 (𝜏)} + 𝑟(𝑛 − 𝑟 + 1)𝐿{𝑝 (𝜏)} = (𝑟 + 1)(𝑛 − 𝑟)𝐿{𝑝 (𝜏)}
𝐿{𝑝 (𝜏)} 𝑠 + 𝑟(𝑛 − 𝑟 + 1) = (𝑟 + 1)(𝑛 − 𝑟)𝐿{𝑝 (𝑡)}
(𝑟 + 1)(𝑛 − 𝑟)𝐿{𝑝 (𝜏)}
𝐿{𝑝 (𝜏)} = (7)
𝑠 + 𝑟 (𝑛 − 𝑟 + 1 )
Let r=n-1
(𝑛 − 1 + 1)(𝑛 − 𝑛 + 1)𝐿{𝑝 (𝜏)}
𝐿{𝑝 (𝜏)} =
𝑠 + (𝑛 − 1)(𝑛 − 𝑛 + 1 + 1)
(𝑛)(1)𝐿{𝑝 (𝜏)}
𝐿{𝑝 (𝜏)} =
𝑠 + (𝑛 − 1)(2)
𝑛𝐿{𝑝 (𝜏 )}
𝐿{𝑝 (𝜏)} =
(𝑠 + 2(𝑛 − 1))
𝑛 1
𝐿{𝑝 (𝜏)} = ∗
(𝑠 + 2(𝑛 − 1)) 𝑠 + 𝑛
1. 𝑛
𝐿{𝑝 (𝜏)} = (8)
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))
For r=n-2
(n − 2 + 1)(𝑛 − n + 2)𝐿{𝑝 (𝜏)}
𝐿{𝑝 (𝜏)} =
𝑠 + (𝑛 − 2)(𝑛 − 𝑛 + 2 + 1)
(n − 1)(2)𝐿{𝑝 (𝜏)}
=
𝑠 + (𝑛 − 2)(3)
2(n − 1)𝐿{𝑝 (𝜏)}
=
(𝑠 + 3(𝑛 − 2))
2(n − 1) 𝑛
𝐿{𝑝 (𝜏)} = ∗
(𝑠 + 3(𝑛 − 2)) (𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))
Page 7 of 15
1.2𝑛(n − 1)
𝐿{𝑝 (𝜏)} = (9)
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))
For r=n-3
3(n − 2)
𝐿{𝑝 (𝜏)} = 𝐿{𝑝 (𝜏)}
(𝑠 + 4(𝑛 − 3))
1.2.3 𝑛(n − 1)(𝑛 − 2)
𝐿{𝑝 (𝜏 )} =
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3))
For r=n-(n-r)
𝐿 𝑝 ( ) (𝜏 )
= 𝐿{𝑝 (𝜏 )}
1.2.3 … (𝑛 − 𝑟)𝑛(n − 1)(𝑛 − 2) … … (𝑛 − (𝑛 − 𝑟 − 1))
=
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)(𝑛 − 𝑛 + 𝑟))
1.2.3 … (𝑛 − 𝑟 + 1)(𝑛 − 𝑟)𝑛(n − 1)(𝑛 − 2) … … (𝑛 − (𝑛 − 𝑟 − 1))
𝐿{𝑝 (𝜏)} =
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
(𝑛 − 𝑟)(𝑛 − 𝑟 + 1) … .3.2.1𝑛(n − 1)(𝑛 − 2) … … (𝑛 − (𝑛 − 𝑟 − 1))
=
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
r!=r(r-1)(r-2)…..3.2.1
(𝑛 − 𝑟)! 𝑛(n − 1)(𝑛 − 2) … … 𝑛 − (𝑛 − 𝑟 − 1)
𝐿{𝑝 (𝜏 )} =
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
(𝑛 − 𝑟)! 𝑛(n − 1)(𝑛 − 2) … … (𝑟 + 1)
=
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
(𝑛 − 𝑟)! 𝑛(n − 1)(𝑛 − 2) … … (𝑟 + 1)
=
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
1
𝐿 =𝑒
𝑠+𝑎
Multiplying and dividing by r!
𝐿{𝑝 (𝑡)}
(𝑛 − 𝑟)! 𝑛(n − 1)(𝑛 − 2) … … 𝑟. 𝑟!
=𝛽
𝑟! (𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
(𝑛 − 𝑟)! 𝑛(n − 1)(𝑛 − 2) … … (𝑟 + 1)𝑟!
=
𝑟!
1
∗
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
Page 8 of 15
(𝑛 − 𝑟)! 𝑛(n − 1)(𝑛 − 2) … … (𝑟 + 1)𝑟(𝑟 − 1) … . .3.2.1
=
𝑟!
1
∗
(𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
(𝑛 − 𝑟)! 𝑛! 1
=
𝑟! (𝑠 + 𝑛)(𝑠 + 2(𝑛 − 1))(𝑠 + 3(𝑛 − 2))(𝑠 + 4(𝑛 − 3)) … … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟
(𝑛 − 𝑟)! 𝑛! 1
𝐿{𝑝 (𝑡)} = (10)
𝑟! (𝑠 + 𝑗(𝑛 − (𝑗 − 1)
1
𝑓 (𝑥 )𝑑𝑥 = 𝑑𝑥
(𝑥 + 1)(𝑥 + 2)(𝑥 + 3)
1 𝐴 𝐵 𝐶
= + +
(𝑥 + 1)(𝑥 + 2)(𝑥 + 3) (𝑥 + 1) (𝑥 + 2) (𝑥 + 3)
1 = 𝐴(𝑥 + 2)(𝑥 + 3) + 𝐵(𝑥 + 3)(𝑥 + 1) + 𝐶(𝑥 + 1)(𝑥 + 2)
0𝑥 + 0𝑥 + 1 = 𝑥 (𝐴 + 𝐵 + 𝐶) + 𝑥(5𝐴 + 4𝐵 + 3𝐶) + (6𝐴 + 3𝐵 + 2𝐶)
⟹ 𝐴 + 𝐵 + 𝐶 = 0, 5𝐴 + 4𝐵 + 3𝐶, 6𝐴 + 3𝐵 + 2𝐶
After solving above three equations we will get
1
𝐴= , 𝐵 = −1, 𝐶 = 1/2
2
1/2 1 1/2
= − + 𝑑𝑥
(𝑥 + 1) (𝑥 + 2) (𝑥 + 3)
(𝑛 − 𝑟)! 𝑛! 1
𝐿{𝑝 (𝜏)} =
𝑟! (𝑠 + 𝑗(𝑛 − (𝑗 − 1)
Let
(𝑛 − 𝑟)! 𝑛! 1
𝐿{𝑝 (𝜏)} =
𝑟! (𝑠 + 𝑗(𝑛 − (𝑗 − 1)
𝐶, 𝐶, 𝐶,
𝐿{𝑝 (𝜏)} = + + ⋯..+ (12)
𝑠 + 𝑛 𝑠 + 2(𝑛 − 1) 𝑠 + 𝑟(𝑛 − 𝑟 + 1)
Then say,
Page 9 of 15
lim (𝑠 + 𝑘(𝑛 − 𝑘 + 1) 𝐿{𝑝 (𝜏)} = 𝐶 ,
→ ( )
(𝑛 − 𝑟)! 𝑛! 1
lim =𝐶 ,
→ ( ) 𝑟! (𝑠 + 𝑗(𝑛 − (𝑗 − 1)
(𝑛 − 𝑟)! 𝑛! 1
𝐶 , = lim (𝑠 + 𝑘(𝑛 − 𝑘 + 1))
→ ( ) 𝑟! (𝑠 + 𝑗(𝑛 − (𝑗 − 1)
(𝑛 − 𝑟)! 𝑛!
𝐶 , =
𝑟!
⎡ ⎤
⎢ ⎥
⎢ ⎥
⎢ 1 ⎥
lim (𝑠 + 𝑘(𝑛 − 𝑘 + 1)) ⎢ ⎥
→ ( ) (𝑠 + 𝑛) 𝑠 + 2(𝑛 − 1) 𝑠 + 3(𝑛 − 2) 𝑠 + 4(𝑛 − 3)
⎢ ⎥
⎢… 𝑠 + (𝑘 − 1) 𝑛 − (𝑘 − 2) 𝑠 + (𝑘) 𝑛 − (𝑘 − 1) ⎥
⎢ ⎥
⎣ 𝑠 + (𝑘 + 1)(𝑛 − 𝑘)) … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟 ⎦
(𝑛 − 𝑟)! 𝑛!
=
𝑟!
⎡ ⎤
⎢ ⎥
⎢ ⎥
⎢ 1 ⎥
lim ⎢ ⎥
→ ( ) (𝑠 + 𝑛) 𝑠 + 2(𝑛 − 1) 𝑠 + 3(𝑛 − 2) 𝑠 + 4(𝑛 − 3)
⎢ ⎥
⎢ … 𝑠 + (𝑘 − 1) 𝑛 − (𝑘 − 2) ⎥
⎢ ⎥
⎣ 𝑠 + (𝑘 + 1)(𝑛 − 𝑘)) … . . (𝑠 + (𝑛 − 𝑟 + 1)𝑟 ⎦
⎡ ⎤
⎢ ⎥
⎢ ⎥
⎢ ⎥
(𝑛 − 𝑟)! 𝑛! ⎢ 1 ⎥
= lim ⎢ ⎥
𝑟! → ( ) (𝑠 + 𝑛) 𝑠 + 2(𝑛 − 1) 𝑠 + 3(𝑛 − 2) 𝑠 + 4(𝑛 − 3)
⎢ ⎥
⎢ … 𝑠 + (𝑘 − 1) 𝑛 − (𝑘 − 2) ⎥
⎢ ⎥
⎢ 𝑠 + (𝑘 + 1)(𝑛 − 𝑘)) 𝑠 + (𝑘 + 2)(𝑛 − 𝑘 + 1)) … . . (𝑠 + (𝑟 − 1)(𝑛 − 𝑟)⎥
⎣ (𝑠 + (𝑛 − 𝑟 + 1)𝑟) ⎦
Let,
Page 10 of 15
𝐴
1
= lim
→ ( )
(𝑠 + 𝑛) 𝑠 + 2(𝑛 − 1) 𝑠 + 3(𝑛 − 2) 𝑠 + 4(𝑛 − 3) … 𝑠 + (𝑘 − 1) 𝑛 − (𝑘 − 2)
𝐵
⎡ ⎤
⎢ 1 ⎥
= lim
→ ( ) ⎢ 𝑠 + (𝑘 + 1)(𝑛 − 𝑘)) 𝑠 + (𝑘 + 2)(𝑛 − 𝑘 + 1)) … . . 𝑠 + (𝑟 − 1)(𝑛 − 𝑟 ) ⎥
⎢ ⎥
⎣ (𝑠 + (𝑛 − 𝑟 + 1)𝑟) ⎦
1
𝐴=
(−𝑘(𝑛 − 𝑘 + 1) + 𝑛) −𝑘(𝑛 − 𝑘 + 1) + 2(𝑛 − 1) −𝑘(𝑛 − 𝑘 + 1) + 3(𝑛 − 2)
−𝑘(𝑛 − 𝑘 + 1) + 4(𝑛 − 3) … −𝑘(𝑛 − 𝑘 + 1) + (𝑘 − 1) 𝑛 − (𝑘 − 2)
1
=
−𝑘(𝑛 − 𝑘 + 1) + 1. 𝑘 + 1(𝑛 − 𝑘) (−𝑘 (𝑛 − 𝑘 + 1) + 2𝑘 + 2(𝑛 − 𝑘 ) − 2)
(−𝑘 (𝑛 − 𝑘 + 1) + 3𝑘 + 3(𝑛 − 𝑘) − 6)
(−𝑘 (𝑛 − 𝑘 + 1) + 4𝑘 + 4(𝑛 − 𝑘 ) − 12) … . .
(−𝑘(𝑛 − 𝑘 + 1) + (𝑘 − 1)𝑘 + (𝑘 − 1)(𝑛 − 𝑘) − (𝑘 − 1)𝑘)
1
=
{−(𝑛 − 𝑘)(𝑘 − 1)}{−(𝑛 − 𝑘)(𝑘 − 2) − 𝑘 + 2𝑘 − 2}
{−(𝑛 − 𝑘 )(𝑘 − 3) − 𝑘 + 3𝑘 − 6}
{−(𝑛 − 𝑘)(𝑘 − 4) − 𝑘 + 4𝑘 − 12} … . .
{(𝑛 − 𝑘)(−𝑘 + 𝑘 − 1)}
1
=
{−(𝑛 − 𝑘 (𝑘 − 1)}{− 𝑛 − 𝑘)(𝑘 − 2) + (𝑘 − 2)}
) (
{−(𝑛 − 𝑘 )(𝑘 − 3) + 2(𝑘 − 3)}
{−(𝑛 − 𝑘 )(𝑘 − 4) + 3(𝑘 − 4)} … . .
{(𝑛 − 𝑘)(−1)}
1
=
{−(𝑛 − 𝑘)(𝑘 − 1)}{−(𝑛 − 𝑘 − 1)(𝑘 − 2)}
{−(𝑛 − 𝑘 − 2)(𝑘 − 3)}
{−(𝑛 − 𝑘 − 4)(𝑘 − 4)} … . .
{−(𝑛 − 𝑘 )(1)}
(−1)
𝐴=
{(𝑛 − 𝑘)(𝑛 − 𝑘 − 1)(𝑛 − 𝑘 − 2)(𝑛 − 𝑘 − 3)(𝑛 − 𝑘 − 4) … . (𝑛 − 𝑘)}
{(𝑘 − 1)(𝑘 − 2)(𝑘 − 3) … .2.1}
(−𝟏)𝒌 𝟏
𝑨=
(𝒏 − 𝒌)! (𝒌 − 𝟏)!
Page 11 of 15
Similarly
𝐵
⎡ ⎤
⎢ 1 ⎥
= lim
→ ( ) ⎢ 𝑠 + (𝑘 + 1)(𝑛 − 𝑘)) 𝑠 + (𝑘 + 2)(𝑛 − 𝑘 + 1)) … . . 𝑠 + (𝑟 − 1)(𝑛 − 𝑟 ) ⎥
⎢ ⎥
⎣ (𝑠 + (𝑛 − 𝑟 + 1)𝑟) ⎦
Multiply and divide by (n-2k+1) and (r-k-1)! We will get
(n − 2k + 1) (r − k − 1)!
𝐵=
(𝑛 − 𝑟 − 𝑘 + 1 )!
Hence
(𝑛 − 𝑟)! 𝑛!
𝐶 , = ∗𝐴∗𝐵
𝑟!
(𝑛 − 𝑟)! 𝑛! (−1) (n − 2k + 1) (r − k − 1)!
𝐶 , =
𝑟! (n − k)! (k − 1)! (𝑛 − 𝑟 − 𝑘 + 1 )!
(𝑛 − 𝑟)! 𝑛! (−1) (n − 2k + 1) (r − k − 1)!
𝐶 , =
𝑟! (n − k)! (k − 1)! (𝑛 − 𝑟 − 𝑘 + 1)!
(𝑛 − 𝑟)! 𝑛! 1
lim =𝐶 , (13)
𝑟! → ( ) (𝑠 + 𝑗(𝑛 − (𝑗 − 1)
(𝑛 − 𝑟)! 𝑛! (−1) (𝑛 − 2𝑘 + 1)(𝑟 − 𝑘 + 1)!
𝐶 , =
𝑟! (𝑘 − 1)! (𝑛 − 𝑘 )! (𝑛 − 𝑟 − 𝑘 + 1)!
𝐶, 𝐶, 𝐶,
𝐿{𝑝 (𝜏)} = + + ⋯..+ (11)
𝑠 + 𝑛 𝑠 + 2(𝑛 − 1) 𝑠 + 𝑟(𝑛 − 𝑟 + 1)
Taking inverse Laplace transformation
𝐶, 𝐶, 𝐶,
𝑝 (𝜏 ) = 𝐿 + + ⋯..+
𝑠 + 𝑛 𝑠 + 2(𝑛 − 1) 𝑠 + 𝑟(𝑛 − 𝑟 + 1)
𝐶, 𝐶, 𝐶,
=𝐿 +𝐿 + ⋯ . . +𝐿
𝑠+𝑛 𝑠 + 2(𝑛 − 1) 𝑠 + 𝑟(𝑛 − 𝑟 + 1)
( ) ( )
=𝐶 , 𝑒 +𝐶 , 𝑒 + ⋯ . . +𝐶 , 𝑒
Page 12 of 15
( )
𝑝 (𝜏 ) = 𝐶, 𝑒
( )
𝑝 (𝑡) = 𝐶, 𝑒
Probability of r susceptible in time t
( )
𝑝 (𝑡 ) = 𝐶, 𝑒 𝑖𝑓 𝑟 > 𝑛/2
Here
(𝑛 − 𝑟)! 𝑛! (−1) (𝑛 − 2𝑘 + 1)(𝑟 − 𝑘 + 1)
𝐶 , =
𝑟! (𝑘 − 1)! (𝑛 − 𝑘 )! (𝑛 − 𝑟 − 𝑘 + 1)
Duration of an Epidemic Model:
Duration of an epidemic model is the time that elapses before all susceptible becomes
infected.
Let there are j infectives and (n-j+1) susceptible, the chance of a new infection during an
infinitesimal interval ∆𝑡 is given by
𝑗(𝑛 − 𝑗 + 1)∆𝑡
The interval 𝜏 between the occurrence of the jth infection and (j+1)th infection has a
negative exponential distribution given by
( )
𝑓 𝜏 = 𝑗(𝑛 − 𝑗 + 1)𝑒 (14)
It is clear that the r.v. 𝜏 ′𝑠 are all independently distributed. The duration of time T of the
epidemic is given by
𝑇= 𝜏
Now
𝐸(𝑒 ) = 𝜆 𝑒 𝑒 𝑑𝑥
( )
=𝜆 𝑒 𝑑𝑥
Page 13 of 15
𝑒 ( )
=𝜆 −
(𝜆 − 𝑡)
(𝑒 − 𝑒 )
=𝜆 −
(𝜆 − 𝑡)
𝜆
=
(𝜆 − 𝑡)
1
𝐸 (𝑒 ) = 𝑡 (15)
1−
𝜆
Taking log
1
log 𝐸(𝑒 ) = 𝑙𝑜𝑔 𝑡
1−
𝜆
𝑡
= 𝑙𝑜𝑔(1) − 𝑙𝑜𝑔 1 −
𝜆
𝑡
= −𝑙𝑜𝑔 1 −
𝜆
𝑡 1 𝑡 1 𝑡
𝐸 (𝑒 ) = + + ⋯..+ + ⋯. (16)
𝜆 2 𝜆 𝑟 𝜆
𝑡 1 𝑡 2! 𝑡 (𝑟 − 1)! 𝑡
=− − − − …..− − ⋯..
𝜆 2! 𝜆 3! 𝜆 𝑟! 𝜆
𝑡 1 𝑡 2! 𝑡 (𝑟 − 1)! 𝑡
= + + …..+ + ⋯ ..
𝜆 2! 𝜆 3! 𝜆 𝑟! 𝜆
Now rth cumulant of a negative exponential variate with parameter 𝜆
𝑡 (𝑟 − 1)!
𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 = (17)
𝑟! 𝜆
rth cumulant of a negative exponential variate Tj with parameter 𝑗(𝑛 − 𝑗 + 1)
𝑡 (𝑟 − 1)!
𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑜𝑓 =
𝑟! 𝑗 (𝑛 − 𝑗 + 1)
Sum of rth cumulant of 𝑇
(𝑟 − 1)!
𝑘 = (18)
𝑗 (𝑛 − 𝑗 + 1)
Page 14 of 15
For small n 𝑘 can be calculated directly by using (18).
Page 15 of 15