get.E.vacc: Calculates the expected number of vaccinations for all...
In vacem: Vaccination Activities Coverage Estimation Model

Description Usage Arguments Details Value Note Note

Calculates the expected number of vaccinations for all individuals in a group of observations and under a described set of vaccination activities (e.g. campaigns).

1 2	get.E.vacc(obs, camps, trans.rho, alpha, z, w, cdf.fun, cond.on.accessible=FALSE, ...)

`obs`	a data frame with one row per observation and columns for @c date of observation and @c age of individual on that date; <i>Note: all ages should be in months</i>
`camps`	a data frame with one row per activity and columns providing the activity's @c date, targeted age range (@c age.low and @c age.high), size of targeted population (@c N) and the number of vaccine doses nominally distributed (@c v)
`trans.rho`	the log-odds (aka logit) of @c rho, the proportion of the population (scalar) that can be vaccinated; see note above
`alpha`	the vaccination efficiency parameter (scalar) for the campaign; @codealpha = -inf implies perfect efficiency, @codealpha = 0 implies random efficiency; see note above
`z`	the eligibility matrix for each individual/campaign pairing, i.e. @codez[i,j] = 1 if this individual @c i is eligible for campaign @c j, otherwise @codez[i,j] = 0; <i>Note: if @c z is @c NULL, then the eligibilty matrix is calculated from @p obs and @p camps parameters using @code z.matrix function.</i>
`w`	the weight matrix for each individual/campaign pairing, i.e. @code w[i,k] gives the probability of individual @c i receiving a routine vaccination in year @c k <i>Note: if @c w is @c NULL, then the weight matrix is calculated from @p obs, @p camps, @p cdf.fun and @p ... parameters using @code w.matrix function.</i>
`cdf.fun`	a cumulative distribution function (CDF) used to calculate the probability of routine vaccination based upon age, e.g. @code constant.post.9mo . <i>Note: this parameter is only required if the weight matrix, @p w, is <u>not</u> provided.</i>
`cond.on.accessible`	a flag indicating whether the return values should be conditioned on the individual's inclusion in the accessible population, i.e. @p rho; the default is @c FALSE
`...`	any additional arguments to the @p cdf.fun, e.g. @c lambda <i>Note: this parameter is only required if the weight matrix, @p w, is not provided.</i>

The @c get.E.vacc function simply applies the @code E.vacc function to all the entries in observations data frame, @p obs. The @p obs argument should contain the immunization information as derived, for example, from a Demographic and Health Survey (DHS). The @c get.E.vacc function uses the observations and vaccination activity descriptions, @p camps, to calculate the eligibility matrix, @code z.matrix and the weight matrix @code w.matrix if not provided. Then the expected vaccination doses are calculated by calling @c E.vacc with each individual's eligibility and weight vector as well as the relevant vaccination activity information (i.e. @p camps$N and @p camps$v), the efficiency parameter (@p alpha), the accessibility multiplier (@c rho) and conditionality flag (@p cond.on.accessible).

a vector containing the expected number of vaccinations for each individual in observations data frame, @p obs.

@see E.vacc @see z.matrix @see w.matrix @see nine.month.pointmass @see constant.post.9mo

The @p trans.rho argument represents the @linklogit of @c rho, i.e. @code trans.rho = log( rho / (1-rho) ) , where @c rho is the proportion of population that is accessible to any vaccination activity and @code 1-rho the inaccessible portion. It follows then that @c rho is calculated as the inverse-logit (or logistic function) of @p trans.rho, that is: @code rho <- exp( trans.rho ) / ( 1 + exp(trans.rho) )

The inefficiency measure @p alpha is defined as the natural log of psi, i.e. @codepsi = e^alpha. @codepsi = 0 (@codealpha = -inf) represents perfect efficiency, i.e. when every dose results in an additional vaccinee. @codepsi = 1 (@codealpha = 0) represents random efficiency, i.e. when probability of any dose resulting in an additional vaccinee is independent of receiving a dose previously during the same campaign.

vacem documentation built on April 14, 2017, 7:28 p.m.