rates: Conversion of counts to rates

In akmedoids: Anchored Kmedoids for Longitudinal Data Clustering

Description

Calculates rates from 'observed' count and a denominator data

Usage

rates(traj, denomin, id_field, multiplier)

Arguments

traj	[matrix (numeric)] longitudinal (e.g. observed count) data (m x n). Each row represents an individual trajectory (of observations). The columns show the observations at consecutive time steps.
denomin	[matrix (numeric)] longitudinal (denominator) data of the same column as 'traj' (n).
id_field	[numeric or character] Default is TRUE. The first column of both the 'traj' and the 'denomin' object must be the unique (id) field. If FALSE, the function will terminate. The assumption is that columns of both the traj and denominat corresponds. That is, column2, column3, ... represent time points 2, 3, ..., respectively, in each object.
multiplier	[numeric] A quantify by which to the ratio traj/denomin is expressed. Default is 100.

Value

An object which comprised of four output variables, namely: (i) '$common_ids' - individual ids present in both 'traj' (trajectory data) and 'denomin' (denominator data); (ii) '$ids_unique_to_traj_data' - individual ids unique to trajectory data (i.e. not present in the denominator data); (iii) '$ids_unique_to_denom_data' - individual ids unique to denominator data (i.e. not present in the trajectory data); (iv) “ - a dataframe of rates estimates. Note: only the individual ids in '$rates_estimates' are used in the 'rates' estimation.

Examples

traj2 <- data_imputation(traj, id_field = TRUE, method = 2,
replace_with = 1, fill_zeros = FALSE)

pop <- popl #read denominator data

pop2 <- as.data.frame(matrix(0, nrow(popl), ncol(traj)))

colnames(pop2) <- names(traj2$CompleteData)

pop2[,1] <- as.vector(as.character(pop[,1]))

pop2[,4] <- as.vector(as.character(pop[,2]))

pop2[,8] <- as.vector(as.character(pop[,3]))

list_ <- c(2, 3, 5, 6, 7, 9, 10) #vector of missing years

#fill the missing fields with 'NA'
for(u_ in seq_len(length(list_))){
    pop2[,list_[u_]] <- "NA"
}

#estimate missing fields
pop_imp_result <- data_imputation(pop2, id_field = TRUE, method = 2,
replace_with = 1, fill_zeros = FALSE)

#calculate rates i.e. crimes per 200 population
crime_rates <- rates(traj2$CompleteData, denomin=pop_imp_result$CompleteData,
id_field=TRUE, multiplier = 200)

akmedoids documentation built on April 13, 2021, 9:07 a.m.