Parameters for the "lancelot" model.
lancelot_parameters(
start_date,
region,
beta_date = NULL,
beta_value = NULL,
beta_type = "piecewise-linear",
population = NULL,
carehome_beds = NULL,
severity = NULL,
progression = NULL,
observation = NULL,
sens_and_spec = NULL,
initial_seed_size = 30,
initial_seed_pattern = 1,
eps = 0.1,
m_CHW = 4e-06,
m_CHR = 5e-05,
strain_transmission = 1,
strain_seed_date = NULL,
strain_seed_size = NULL,
strain_seed_pattern = NULL,
strain_rel_gamma_E = 1,
strain_rel_gamma_A = 1,
strain_rel_gamma_P = 1,
strain_rel_gamma_C_1 = 1,
strain_rel_gamma_C_2 = 1,
strain_rel_gamma_PCR_pre = 1,
strain_rel_gamma_PCR_pos = 1,
strain_rel_p_sympt = 1,
strain_rel_p_hosp_if_sympt = 1,
strain_rel_p_icu = 1,
strain_rel_p_death = 1,
strain_rel_p_G_D = 1,
rel_susceptibility = 1,
rel_p_sympt = 1,
rel_p_hosp_if_sympt = 1,
rel_p_death = 1,
rel_infectivity = 1,
vaccine_progression_rate = NULL,
vaccine_schedule = NULL,
vaccine_index_dose2 = NULL,
vaccine_index_booster = NULL,
vaccine_catchup_fraction = 1,
n_doses = 2L,
vacc_skip_progression_rate = NULL,
vacc_skip_to = NULL,
vacc_skip_weight = NULL,
waning_rate = 0,
exp_noise = 1e+06,
cross_immunity = 1
)The start date, as a sircovid_date() (i.e.,
the number of days into 2020)
The region to run the model for. This will bee used to get population data, which is currently fixed within the package and is limited to "uk", the four constituent nations ("england", "wales", "scotland", "northern_ireland") and the 7 NHS regions (e.g., "midlands"). These names are case insensitive.
A vector of dates (each as a sircovid_date())
for changes in beta (the contact rate parameter), or NULL if
a single value is used for all times (see
sircovid_parameters_piecewise_linear() or
sircovid_parameters_piecewise_constant(), where
this is passed as date). If beta_type = "piecewise-constant",
then the first date must be 0.
A vector of values for beta (the contact rate
parameter). If not given, and if beta_date is NULL then a
value of 0.1 will be used through the whole simulation,
otherwise if beta_date is NULL this must be a scalar. If
beta_date is given then beta_date and beta_value must have
the same length (see sircovid_parameters_piecewise_linear() or
sircovid_parameters_piecewise_constant(), where this is passed
as value).
The type of form used for beta (the contact rate
parameter), which currently can be "piecewise-linear" or
"piecewise-constant"
Population data. A vector of length 17 for the population
size for age groups 0-4, 5-9, ..., 75-79, 80+. If NULL, the population
data will be sourced within the package for the specified region (only
if available).
The number of care home beds in the region. If NULL,
this will be sourced within the package for the specified region (only
if available).
Severity data, via Bob Verity's markovid
package. This needs to be NULL (use the default bundled data
version in the package), a data.frame object (for raw severity
data) or a list (for data that has already been processed by
sircovid for use). New severity data comes from Bob Verity
via the markovid package, and needs to be carefully calibrated
with the progression parameters.
Progression data
Either NULL or a list of observation parameters. If
NULL, then a list of observation parameters will be generated using
lancelot_parameters_observation(exp_noise)
Either NULL or a list of diagnostic test sensitivity
and specificity parameters. If NULL, then a list of sensitivity and
specificity parameters will be generated using
lancelot_parameters_sens_and_spec()
Initial size of seeding from the S to E
compartment; all seeding is in the 15-19 year old group from
start_date according to initial_seed_pattern. The default is 30.
A vector of seeding weights for the initial
seeding. The length represents the number of steps to seed over from the
start_date, and the initial_seed_size is split over these steps
according to those weights. If start_date is not a multiple of the step
size (and thus falls between two steps) then we weight over an additional
step and adjust the weights according to how far the start_date is from
the previous full step.
Change in contact rate for carehome residents
Contact rate between carehome workers and either residents or workers
Contact rate between carehome residents
Vector of length two for relative transmissibility of each strain modelled. Length will define the number of strains used in the model, either 1 or 2.
Either NULL (no seeding) or a
sircovid_date corresponding to the date the seeding of strain
2 begins.
Either NULL (no seeding) or the size of strain 2
seeding from the S to E compartment; all seeding is in the 15-19 year old
group from strain_seed_date according to strain_seed_pattern.
Either NULL (no seeding) or a vector of seeding
weights for the initial seeding. The length represents the number of steps
to seed over from the start_date, and the strain_seed_size is split
over these steps according to those weights. If start_seed_date is not a
multiple of the step size (and thus falls between two steps) then we
weight over an additional step and adjust the weights according to how far
the strain_seed_date is from the previous full step.
Vector of relative rates of progression out of
E (gamma_E) for each
strain modelled. If 1 all strains have same rates. Otherwise vector of
same length as strain_transmission, with entries that determines the
relative scaling of the defaults for each strain.
Vector of relative rates of progression out of
I_A (gamma_A) for each
strain modelled. If 1 all strains have same rates. Otherwise vector of
same length as strain_transmission, with entries that determines the
relative scaling of the defaults for each strain.
Vector of relative rates of progression out of
I_P (gamma_P) for each
strain modelled. If 1 all strains have same rates. Otherwise vector of
same length as strain_transmission, with entries that determines the
relative scaling of the defaults for each strain.
Vector of relative rates of progression out of
I_C_1 (gamma_C_1) for each
strain modelled. If 1 all strains have same rates. Otherwise vector of
same length as strain_transmission, with entries that determines the
relative scaling of the defaults for each strain.
Vector of relative rates of progression out of
I_C_2 (gamma_C_2) for each
strain modelled. If 1 all strains have same rates. Otherwise vector of
same length as strain_transmission, with entries that determines the
relative scaling of the defaults for each strain.
Vector of relative rates of PCR
pre-positive duration for each strain modelled. if '1' all strains have
same rates. Otherwise vector of same length as strain_transmission, with
entries that determines the relative scaling of the defaults for each
strain.
Vector of relative rates of PCR positivity
duration for each strain modelled. if '1' all strains have same rates.
Otherwise vector of same length as strain_transmission, with entries
that determines the relative scaling of the defaults for each strain.
Vector of relative probabilities of
symptoms for
each strain modelled. If 1 all strains have same
probabilities of symptoms. Otherwise vector of same length as
strain_transmission, where the first value should be 1 (for the first
strain) and subsequent values between 0 and 1. In this case parameters
will be "mirrored" for pseudostrains i.e. the relative probability of
symptoms will be assume the same irrespective of previous infection
with another strain. Alternatively, a vector of twice the length of
strain_transmission can be provided to allow specifying directly
relative probability of symptoms for each pseudostrain
(with strain 3 - 1.2 and strain 4 = 2.1). To ensure valid
probabilities, p_sympt is upper-truncated at 1 after scaling.
Vector of relative probabilities of
hospitalisation given symptoms for
each strain modelled. If 1 all strains have same
probabilities of hospitalisation. Otherwise vector of same length as
strain_transmission, where the first value should be 1 (for the first
strain) and subsequent values between 0 and 1. In this case parameters
will be "mirrored" for pseudostrains i.e. the relative probability of
hospitalisation will be assume the same irrespective of previous infection
with another strain. Alternatively, a vector of twice the length of
strain_transmission can be provided to allow specifying directly
relative probability of hospitalisation for each pseudostrain
(with strain 3 - 1.2 and strain 4 = 2.1). To ensure valid
probabilities, p_hosp_if_sympt is upper-truncated at 1 after scaling.
Vector of relative probabilities of
icu given hospitalised for
each strain modelled. If 1 all strains have same
probabilities of icu admission. Otherwise vector of same length as
strain_transmission, where the first value should be 1 (for the first
strain) and subsequent values between 0 and 1. In this case parameters
will be "mirrored" for pseudostrains i.e. the relative probability of
icu admission will be assume the same irrespective of previous infection
with another strain. Alternatively, a vector of twice the length of
strain_transmission can be provided to allow specifying directly
relative probability of icu admission for each pseudostrain
(with strain 3 - 1.2 and strain 4 = 2.1). To ensure valid
probabilities, p_icu is upper-truncated at 1 after scaling.
Vector of relative probabilities of death for
each strain modelled. If 1 all strains have same
probabilities of death. Otherwise vector of same length as
strain_transmission, where the first value should be 1 (for the first
strain) and subsequent values between 0 and 1. In this case parameters
will be "mirrored" for pseudostrains i.e. the relative probability of
death will be assume the same irrespective of previous infection with
another strain. Alternatively, a vector of twice the length of
strain_transmission can be provided to allow specifying directly
relative probability of death for each pseudostrain (with strain 3 - 1.2
and strain 4 = 2.1). To ensure valid
probabilities, p_death is upper-truncated at 1 after scaling.
Vector of relative probabilities of death in the
coummunity for each strain modelled. If 1 all strains have same
probabilities of death. Otherwise vector of same length as
strain_transmission, where the first value should be 1 (for the first
strain) and subsequent values between 0 and 1. In this case parameters
will be "mirrored" for pseudostrains i.e. the relative probability of
death will be assume the same irrespective of previous infection with
another strain. Alternatively, a vector of twice the length of
strain_transmission can be provided to allow specifying directly
relative probability of death for each pseudostrain (with strain 3 - 1.2
and strain 4 = 2.1). To ensure valid
probabilities, p_G_D is upper-truncated at 1 after scaling.
A vector or array of values representing the relative susceptibility of individuals in different vaccination groups. If a vector, the first value should be 1 (for the non-vaccinated group) and subsequent values be between 0 and 1. In that case relative susceptibility will be the same across all age groups within one vaccination category, and will be the same for all pathogen strains. Specifying an array instead of a vector allows different relative susceptibilities by age (first dimension of the array), pathogen strain (second dimension) and vaccination group (third dimension); in that case, the first layer (3rd dimension) of rel_susceptibility should be 1 (for the non-vaccinated group) for the first column (first infection with first strain) and other values between 0 and 1
A vector or matrix of values of same dimension as rel_susceptibility representing the relative probability of symptomatic infection in different vaccination groups. If a vector, the first value should be 1 (for the non-vaccinated group) and subsequent values be between 0 and 1. In that case the relative reduction in probability of symptomatic infection will be the same across all age groups and all strains within one vaccination category. Specifying an array instead of a vector allows different relative reductions in probability of symptomatic infection by age (first dimension of the array), pathogen strain (second dimension) and vaccination group (third dimension); in that case, the first layer of rel_p_sympt should be 1 (for the non-vaccinated group) for the first column (first infection with first strain) and other values between 0 and 1
A vector or array of values of same dimension as rel_susceptibility representing the relative probability of hospitalisation for symptomatic cases in different vaccination groups. If a vector, the first value should be 1 (for the non-vaccinated group) and subsequent values be between 0 and 1. In that case the relative reduction in probability of hospitalisation for symptomatic cases will be the same across all age groups and all strains within one vaccination category. Specifying an array instead of a vector allows different relative reductions in probability of hospitalisation for symptomatic cases by age (first dimension of the array), pathogen strain (second dimension) and vaccination group (third dimension); in that case, the first layer of rel_p_hosp_if_sympt should be 1 (for the non-vaccinated group) for the first column (first infection with first strain) and other values between 0 and 1
A vector or array of values of same dimension as rel_susceptibility representing the relative probability of death for severe cases (either in hospital or the community) in different vaccination groups. If a vector, the first value should be 1 (for the non-vaccinated group) and subsequent values be between 0 and 1. In that case the relative reduction in probability of death for severe cases will be the same across all age groups and all strains within one vaccination category. Specifying an array instead of a vector allows different relative reductions in probability of death for severe cases by age (first dimension of the array), pathogen strain (second dimension) and vaccination group (third dimension); in that case, the first layer of rel_p_death should be 1 (for the non-vaccinated group) for the first column (first infection with first strain) and other values between 0 and 1
A vector or array of values representing the relative infectivity of individuals in different vaccination groups, if they are infected. If a vector, the first value should be 1 (for the non-vaccinated group) and subsequent values be between 0 and 1. In that case relative infectivity will be the same across all age groups and all strains within one vaccination category. Specifying an array instead of a vector allows different relative infectivities by age (first dimension of the array), pathogen strain (second dimension) and vaccination group (third dimension); in that case, the first layer of rel_infectivity should be 1 (for the non-vaccinated group) for the first column (first infection with first strain) and other values between 0 and 1
A vector or matrix of values of same dimension as rel_susceptibility representing the rate of movement between different vaccination classes. If a vector, it should have as many values as vaccination classes, and the same rates of progression will be used for all age groups (the first rate is the vaccination rate and the last of the rates is the rate of returning to the initial vacination class); if a matrix, the element on row i and column j is the rate of progression from the jth vaccination class to the (j+1)th for age group i.
A vaccine_schedule object indicating the people to be vaccinated by group over time
The index to use for the second dose
The indices to use for the booster doses
A value between 0 and 1 indicating the proportion of doses not distributed according to schedule (e.g. because too many people were in the I or H compartments and could not be vaccinated at the scheduled time) that we postpone to a later date. A value of 0 means we do not catch up at all on any missed doses; a value of 1 means we try to catch up for all missed doses. This is set to 1 by default
Number of doses given out, including boosters. Default is 2.
A vector of length equal to the number of
vaccine strata. Represents the base progression rate for "vaccine skip
moves" from each strata. The rate is assumed to be the same across all age
groups. Must be 0 for all j for which vacc_skip[j] = 0. Default is a
vector of 0s
A vector of integers of length equal to the number of
vaccine strata. The value of vacc_skip_to[j] represents the vaccine
stratum an individual in vaccine stratum j can move to in a
"vaccine skip move". Values can be 0 (which mean there is no vaccine skip
move from that stratum), or greater than or equal to j + 2 (we already
account for moves between successive vaccine strata). More than one
vaccine skip move to the same vaccine stratum is not allowed, so all
non-zero values in vacc_skip_to must be unique. Overlapping vaccine
skip moves are also not allowed - if a there is a vaccine skip move from
i to j there can be no vaccine skip moves from stratum k where
i < k < j. For example, having vaccine skip moves from 2 to 4 and 3 to 5
would not be allowed, but moves from 2 to 4 and 4 to 6 would be allowed.
Default is a vector of 0s
A vector of weights of length equal to the number
of vaccine strata. If movement into vacc_skip_to[j] from
vacc_skip_to[j] - 1 is controlled by doses, then the "vaccine skip move"
is too, and the weight represents how much those in strata j are
weighted in dose distribution relative to those in vacc_skip_to - 1.
Must be between 0 and 1. Must be 0 for any j for which
vacc_skip_to = 0. Default is a vector of 0s
A single value or a vector of values representing the rates of waning of immunity after infection; if a single value the same rate is used for all age groups; if a vector of values if used it should have one value per age group.
Rate of exponential noise used in the compare
function - typically set to a large value so that noise is small
but non-zero. If set to Inf then there is no noise in the
observation process (not realistic but useful for testing).
A value or vector of same length as
strain_transmission that controls the amount of immunity conferred by
previous infection with one strain. If a scalar is given
then same level of cross immunity is assumed between both strains.
Otherwise a vector of length two should be provided where the first value
is the relative protection against infection with strain 2
following infection with strain 1 (i.e. while in the R1 compartment),
and vice versa for the second value. Values between 0 and 1 are allowed
with values of 1 (default) indicating complete cross-immunity,
and values of 0 mean no cross-immunity. Modelling
'superinfections' (being exposed to one strain after
recovering from another) can be turned off by setting
cross_immunity = 1.
A list of inputs to the model, many of which are fixed and
represent data. These correspond largely to user() calls
within the odin code, though some are also used in processing
just before the model is run.
region <- "london"
lancelot_parameters(sircovid_date("2020-02-01"), region)
# example set up of vaccination parameters independent of age
# 3 groups: 1) unvaccinated, 2) vaccinated with partial immunity
# 3) fully vaccinated (but with an imperfect vaccine). People return
# to group 1 after a period of time in group 3 to mirror waning immunity
# Assumption: immediately after vaccination susceptibility is reduced by
# 20%, and then by 50% when you reach full effect of the vaccination,
# then susceptibility returns to 100% upon waning of vaccine=induced
# immunity
# effect of vaccination similar across all age groups
rel_susceptibility <- c(1, 0.8, 0.5)
# The vaccine also reduces the risk of symptoms
rel_p_sympt <- c(1, 0.6, 0.3)
# and the risk of hospitalisation for those with symptoms
rel_p_hosp_if_sympt <- c(1, 0.95, 0.95)
# and the risk of death for those with severe disease
rel_p_death <- c(1, 0.9, 0.9)
# The vaccine also reduces infectivity of infected individuals by half
rel_infectivity <- c(1, 0.5, 0.5)
# On average 10000 first doses of vaccine are given every day
# Second doses are given on average 12 weeks after the first dose
# vaccine-induced immunity wanes after a period which is exponentially
# distributed and lasts on average 26 weeks (half a year);
# they are similar across all age groups
vaccine_progression_rate <- c(0, 0, 1/(26*7))
daily_doses <- rep(10000, 365)
mean_days_between_doses <- 12 * 7
n <- vaccine_priority_population(region, uptake = 1)
schedule <- vaccine_schedule_future(
0, daily_doses, mean_days_between_doses, n)
# generate model parameters
p <- lancelot_parameters(
sircovid_date("2020-02-01"), region,
rel_susceptibility = rel_susceptibility,
rel_p_sympt = rel_p_sympt,
rel_p_hosp_if_sympt = rel_p_hosp_if_sympt,
rel_p_death = rel_p_death,
rel_infectivity = rel_infectivity,
vaccine_progression_rate = vaccine_progression_rate,
vaccine_schedule = schedule,
vaccine_index_dose2 = 2)
# vaccination parameters are automatically copied across all age groups#
# (and across strains but here we only have 1 strain which is the 2nd
# dimension here)
p$rel_susceptibility
#> , , 1
#>
#> [,1]
#> [1,] 1
#> [2,] 1
#> [3,] 1
#> [4,] 1
#> [5,] 1
#> [6,] 1
#> [7,] 1
#> [8,] 1
#> [9,] 1
#> [10,] 1
#> [11,] 1
#> [12,] 1
#> [13,] 1
#> [14,] 1
#> [15,] 1
#> [16,] 1
#> [17,] 1
#> [18,] 1
#> [19,] 1
#>
#> , , 2
#>
#> [,1]
#> [1,] 0.8
#> [2,] 0.8
#> [3,] 0.8
#> [4,] 0.8
#> [5,] 0.8
#> [6,] 0.8
#> [7,] 0.8
#> [8,] 0.8
#> [9,] 0.8
#> [10,] 0.8
#> [11,] 0.8
#> [12,] 0.8
#> [13,] 0.8
#> [14,] 0.8
#> [15,] 0.8
#> [16,] 0.8
#> [17,] 0.8
#> [18,] 0.8
#> [19,] 0.8
#>
#> , , 3
#>
#> [,1]
#> [1,] 0.5
#> [2,] 0.5
#> [3,] 0.5
#> [4,] 0.5
#> [5,] 0.5
#> [6,] 0.5
#> [7,] 0.5
#> [8,] 0.5
#> [9,] 0.5
#> [10,] 0.5
#> [11,] 0.5
#> [12,] 0.5
#> [13,] 0.5
#> [14,] 0.5
#> [15,] 0.5
#> [16,] 0.5
#> [17,] 0.5
#> [18,] 0.5
#> [19,] 0.5
#>
p$rel_p_sympt
#> , , 1
#>
#> [,1]
#> [1,] 1
#> [2,] 1
#> [3,] 1
#> [4,] 1
#> [5,] 1
#> [6,] 1
#> [7,] 1
#> [8,] 1
#> [9,] 1
#> [10,] 1
#> [11,] 1
#> [12,] 1
#> [13,] 1
#> [14,] 1
#> [15,] 1
#> [16,] 1
#> [17,] 1
#> [18,] 1
#> [19,] 1
#>
#> , , 2
#>
#> [,1]
#> [1,] 0.6
#> [2,] 0.6
#> [3,] 0.6
#> [4,] 0.6
#> [5,] 0.6
#> [6,] 0.6
#> [7,] 0.6
#> [8,] 0.6
#> [9,] 0.6
#> [10,] 0.6
#> [11,] 0.6
#> [12,] 0.6
#> [13,] 0.6
#> [14,] 0.6
#> [15,] 0.6
#> [16,] 0.6
#> [17,] 0.6
#> [18,] 0.6
#> [19,] 0.6
#>
#> , , 3
#>
#> [,1]
#> [1,] 0.3
#> [2,] 0.3
#> [3,] 0.3
#> [4,] 0.3
#> [5,] 0.3
#> [6,] 0.3
#> [7,] 0.3
#> [8,] 0.3
#> [9,] 0.3
#> [10,] 0.3
#> [11,] 0.3
#> [12,] 0.3
#> [13,] 0.3
#> [14,] 0.3
#> [15,] 0.3
#> [16,] 0.3
#> [17,] 0.3
#> [18,] 0.3
#> [19,] 0.3
#>
p$rel_p_hosp_if_sympt
#> , , 1
#>
#> [,1]
#> [1,] 1
#> [2,] 1
#> [3,] 1
#> [4,] 1
#> [5,] 1
#> [6,] 1
#> [7,] 1
#> [8,] 1
#> [9,] 1
#> [10,] 1
#> [11,] 1
#> [12,] 1
#> [13,] 1
#> [14,] 1
#> [15,] 1
#> [16,] 1
#> [17,] 1
#> [18,] 1
#> [19,] 1
#>
#> , , 2
#>
#> [,1]
#> [1,] 0.95
#> [2,] 0.95
#> [3,] 0.95
#> [4,] 0.95
#> [5,] 0.95
#> [6,] 0.95
#> [7,] 0.95
#> [8,] 0.95
#> [9,] 0.95
#> [10,] 0.95
#> [11,] 0.95
#> [12,] 0.95
#> [13,] 0.95
#> [14,] 0.95
#> [15,] 0.95
#> [16,] 0.95
#> [17,] 0.95
#> [18,] 0.95
#> [19,] 0.95
#>
#> , , 3
#>
#> [,1]
#> [1,] 0.95
#> [2,] 0.95
#> [3,] 0.95
#> [4,] 0.95
#> [5,] 0.95
#> [6,] 0.95
#> [7,] 0.95
#> [8,] 0.95
#> [9,] 0.95
#> [10,] 0.95
#> [11,] 0.95
#> [12,] 0.95
#> [13,] 0.95
#> [14,] 0.95
#> [15,] 0.95
#> [16,] 0.95
#> [17,] 0.95
#> [18,] 0.95
#> [19,] 0.95
#>
p$rel_p_death
#> , , 1
#>
#> [,1]
#> [1,] 1
#> [2,] 1
#> [3,] 1
#> [4,] 1
#> [5,] 1
#> [6,] 1
#> [7,] 1
#> [8,] 1
#> [9,] 1
#> [10,] 1
#> [11,] 1
#> [12,] 1
#> [13,] 1
#> [14,] 1
#> [15,] 1
#> [16,] 1
#> [17,] 1
#> [18,] 1
#> [19,] 1
#>
#> , , 2
#>
#> [,1]
#> [1,] 0.9
#> [2,] 0.9
#> [3,] 0.9
#> [4,] 0.9
#> [5,] 0.9
#> [6,] 0.9
#> [7,] 0.9
#> [8,] 0.9
#> [9,] 0.9
#> [10,] 0.9
#> [11,] 0.9
#> [12,] 0.9
#> [13,] 0.9
#> [14,] 0.9
#> [15,] 0.9
#> [16,] 0.9
#> [17,] 0.9
#> [18,] 0.9
#> [19,] 0.9
#>
#> , , 3
#>
#> [,1]
#> [1,] 0.9
#> [2,] 0.9
#> [3,] 0.9
#> [4,] 0.9
#> [5,] 0.9
#> [6,] 0.9
#> [7,] 0.9
#> [8,] 0.9
#> [9,] 0.9
#> [10,] 0.9
#> [11,] 0.9
#> [12,] 0.9
#> [13,] 0.9
#> [14,] 0.9
#> [15,] 0.9
#> [16,] 0.9
#> [17,] 0.9
#> [18,] 0.9
#> [19,] 0.9
#>
p$rel_infectivity
#> , , 1
#>
#> [,1]
#> [1,] 1
#> [2,] 1
#> [3,] 1
#> [4,] 1
#> [5,] 1
#> [6,] 1
#> [7,] 1
#> [8,] 1
#> [9,] 1
#> [10,] 1
#> [11,] 1
#> [12,] 1
#> [13,] 1
#> [14,] 1
#> [15,] 1
#> [16,] 1
#> [17,] 1
#> [18,] 1
#> [19,] 1
#>
#> , , 2
#>
#> [,1]
#> [1,] 0.5
#> [2,] 0.5
#> [3,] 0.5
#> [4,] 0.5
#> [5,] 0.5
#> [6,] 0.5
#> [7,] 0.5
#> [8,] 0.5
#> [9,] 0.5
#> [10,] 0.5
#> [11,] 0.5
#> [12,] 0.5
#> [13,] 0.5
#> [14,] 0.5
#> [15,] 0.5
#> [16,] 0.5
#> [17,] 0.5
#> [18,] 0.5
#> [19,] 0.5
#>
#> , , 3
#>
#> [,1]
#> [1,] 0.5
#> [2,] 0.5
#> [3,] 0.5
#> [4,] 0.5
#> [5,] 0.5
#> [6,] 0.5
#> [7,] 0.5
#> [8,] 0.5
#> [9,] 0.5
#> [10,] 0.5
#> [11,] 0.5
#> [12,] 0.5
#> [13,] 0.5
#> [14,] 0.5
#> [15,] 0.5
#> [16,] 0.5
#> [17,] 0.5
#> [18,] 0.5
#> [19,] 0.5
#>
# Note that this is only the "base" rate as we fill in the first
# column dynamically based on vaccine_daily_doses
p$vaccine_progression_rate
#> [,1] [,2] [,3]
#> [1,] 0 0 0.005494505
#> [2,] 0 0 0.005494505
#> [3,] 0 0 0.005494505
#> [4,] 0 0 0.005494505
#> [5,] 0 0 0.005494505
#> [6,] 0 0 0.005494505
#> [7,] 0 0 0.005494505
#> [8,] 0 0 0.005494505
#> [9,] 0 0 0.005494505
#> [10,] 0 0 0.005494505
#> [11,] 0 0 0.005494505
#> [12,] 0 0 0.005494505
#> [13,] 0 0 0.005494505
#> [14,] 0 0 0.005494505
#> [15,] 0 0 0.005494505
#> [16,] 0 0 0.005494505
#> [17,] 0 0 0.005494505
#> [18,] 0 0 0.005494505
#> [19,] 0 0 0.005494505
### same example as above BUT assume a different effect of vaccine in the
### first age group
n_groups <- 19
n_strains <- 1
# Assumption: vaccine is twice more effective at reducing susceptibility
# in the first age group
rel_susceptibility_agegp1 <- c(1, 0.4, 0.25)
rel_susceptibility_other_agegp <- c(1, 0.8, 0.5)
rel_susceptibility <- array(NA, dim = c(n_groups, n_strains, 3))
rel_susceptibility[1, , ] <- rel_susceptibility_agegp1
for (i in seq(2, n_groups)) {
rel_susceptibility[i, , ] <- rel_susceptibility_other_agegp
}
rel_susceptibility
#> , , 1
#>
#> [,1]
#> [1,] 1
#> [2,] 1
#> [3,] 1
#> [4,] 1
#> [5,] 1
#> [6,] 1
#> [7,] 1
#> [8,] 1
#> [9,] 1
#> [10,] 1
#> [11,] 1
#> [12,] 1
#> [13,] 1
#> [14,] 1
#> [15,] 1
#> [16,] 1
#> [17,] 1
#> [18,] 1
#> [19,] 1
#>
#> , , 2
#>
#> [,1]
#> [1,] 0.4
#> [2,] 0.8
#> [3,] 0.8
#> [4,] 0.8
#> [5,] 0.8
#> [6,] 0.8
#> [7,] 0.8
#> [8,] 0.8
#> [9,] 0.8
#> [10,] 0.8
#> [11,] 0.8
#> [12,] 0.8
#> [13,] 0.8
#> [14,] 0.8
#> [15,] 0.8
#> [16,] 0.8
#> [17,] 0.8
#> [18,] 0.8
#> [19,] 0.8
#>
#> , , 3
#>
#> [,1]
#> [1,] 0.25
#> [2,] 0.50
#> [3,] 0.50
#> [4,] 0.50
#> [5,] 0.50
#> [6,] 0.50
#> [7,] 0.50
#> [8,] 0.50
#> [9,] 0.50
#> [10,] 0.50
#> [11,] 0.50
#> [12,] 0.50
#> [13,] 0.50
#> [14,] 0.50
#> [15,] 0.50
#> [16,] 0.50
#> [17,] 0.50
#> [18,] 0.50
#> [19,] 0.50
#>
# But vaccine has the same impact on probability of symptoms and
# hospitalisation for the symptomatic across all age groups
rel_p_sympt <- array(rep(rel_p_sympt, each = n_groups),
dim = c(n_groups, n_strains, 3))
rel_p_hosp_if_sympt <-
array(rep(rel_p_hosp_if_sympt, each = n_groups),
dim = c(n_groups, n_strains, 3))
rel_p_death <-
array(rep(rel_p_hosp_if_sympt, each = n_groups),
dim = c(n_groups, n_strains, 3))
# And vaccine has the same impact on onwards infectivity across age groups
rel_infectivity <- array(rep(rel_infectivity, each = n_groups),
dim = c(n_groups, n_strains, 3))
# the period of build-up of immunity is the same for all age groups,
# lasting on average 2 weeks,
# but the first age group loses immunity more quickly
# (on average after 3 months) than the other age groups
# (on average after 6 months)
vaccine_progression_rate <- cbind(0, 0,
c(1 / (13 * 7),
rep( 1 / (26 * 7), n_groups - 1)))
# generate model parameters
p <- lancelot_parameters(
sircovid_date("2020-02-01"), region,
rel_susceptibility = rel_susceptibility,
rel_p_sympt = rel_p_sympt,
rel_p_hosp_if_sympt = rel_p_hosp_if_sympt,
rel_p_death = rel_p_death,
rel_infectivity = rel_infectivity,
vaccine_progression_rate = vaccine_progression_rate,
vaccine_schedule = schedule,
vaccine_index_dose2 = 2)
# TODO: add an example of manually set up vaccine schedule