preparedata_fn: A function to prepare mortality data with stratification...

In BayesMoFo: Bayesian Mortality Forecasting

A function to prepare mortality data with stratification (e.g. by product, sex/gender, country, cause of death etc.)

Description

Construct a 3-dimensional data array (dim 1: strata, dim 2: ages, dim 3: years) from raw data frames/arrays suitable for fitting various Bayesian Stochastic Mortality Models.

Usage

preparedata_fn(
  data_df_array,
  data_matrix = FALSE,
  strat_name = NULL,
  ages = NULL,
  years = NULL,
  round = TRUE
)

Arguments

data_df_array	data (normally death counts or exposures) to be provided either as a data frame or a 3-dimensional array. If providing data frame, it should be structured as (column 1: ages, column 2: years, column 3: death/expo data, column 4: strata); if providing array, it should be as (dim 1: strata, dim 2: ages, dim 3: years). If only one stratum (AP data), either remove column 4 or ensure it contains only one stratum label. Alternatively, one can also provide a 2-dimensional matrix (age\times year) but please set data_matrix=TRUE.
data_matrix	a logical value indicating if a 2-dimensional matrix (age\times year) is supplied.
strat_name	a vector of strings indicating names of each stratum.
ages	a numeric vector indicating which ages to use.
years	a numeric vector indicating which years to use.
round	a logical value indicating whether data entries should be rounded.

Value

A list with components:

data

A 3-dimensional data array (dim 1: strata, dim 2: ages, dim 3: years).

strat_name

A vector of strings describing the names of each stratum.

ages

A numeric vector describing the ages used.

years

A numeric vector describing the years used.

n_strat

A numeric value describing the number of strata.

n_ages

A numeric value describing the number of ages.

n_years

A numeric value describing the number of years.

Examples

##################
##if p>1 (more than 1 strata)
##################

#Input: data.frame
data("data_summarised")
head(data_summarised)

#prepare death data
death<-preparedata_fn(data_summarised[,c("Age","Year","Claim","Product")],
strat_name = c("ACI","DB","SCI"),ages=35:65)
#prepare exposure data
expo<-preparedata_fn(data_summarised[,c("Age","Year","Exposure","Product")],
strat_name = c("ACI","DB","SCI"),ages=35:65)

#visualise data
str(death);str(expo)

#works also if data.frame only contains only 1 stratum
death<-preparedata_fn(data_summarised[,
c("Age","Year","Claim","Product")][data_summarised$Product=="ACI",],ages=35:65)

expo<-preparedata_fn(data_summarised[,
c("Age","Year","Exposure","Product")][data_summarised$Product=="ACI",],ages=35:65)

str(death);str(expo)

#Input: 3D data array
data("dxt_array_product");data("Ext_array_product")
death<-preparedata_fn(dxt_array_product,ages=35:65)
expo<-preparedata_fn(Ext_array_product,ages=35:65)

str(death);str(expo)

##################
##if p=1 (only 1 stratum)
##################

#specifying only one of the strats from the data.frame
death<-preparedata_fn(data_summarised[,c("Age","Year","Claim","Product")],
  strat_name = "ACI",ages=35:65)
expo<-preparedata_fn(data_summarised[,c("Age","Year","Exposure","Product")],
  strat_name = "ACI",ages=35:65)
str(death);str(expo)

#if data.frame only contains 1 strat (4 columns)
death<-preparedata_fn(data_summarised[,c("Age","Year","Claim","Product")]
  [data_summarised$Product=="ACI",],ages=35:65)
expo<-preparedata_fn(data_summarised[,c("Age","Year","Exposure","Product")]
  [data_summarised$Product=="ACI",],ages=35:65)
str(death);str(expo)

#if data.frame only contains 1 strat (3 columns)
death<-preparedata_fn(data_summarised[,c("Age","Year","Claim")]
  [data_summarised$Product=="ACI",],ages=35:65)
expo<-preparedata_fn(data_summarised[,c("Age","Year","Exposure")]
  [data_summarised$Product=="ACI",],ages=35:65)
str(death);str(expo)

#Input: 3D data array
death<-preparedata_fn(dxt_array_product,strat_name="ACI",ages=35:65)
expo<-preparedata_fn(Ext_array_product,strat_name="ACI",ages=35:65)
str(death);str(expo)

#Input: 2D matrix 
death<-preparedata_fn(dxt_array_product["ACI",,],data_matrix=TRUE,ages=35:65)
expo<-preparedata_fn(Ext_array_product["ACI",,],data_matrix=TRUE,ages=35:65)
str(death);str(expo)

BayesMoFo documentation built on Aug. 11, 2025, 1:07 a.m.