vwBaseYear: Datasets on Migration Base Year and Type, and Mortality and...

vwBaseYearR Documentation

Datasets on Migration Base Year and Type, and Mortality and Fertility Age Patterns


Datasets giving information on the baseyear and type of migration for each country. The 2012, 2015, 2017 and 2019 datasets also give information on country's specifics regarding mortality and fertility age patterns.




A data frame containing the following variables:


Numerical Location Code (3-digit codes following ISO 3166-1 numeric standard) - see https://en.wikipedia.org/wiki/ISO_3166-1_numeric.


Country name. Not used by the package.


UN internal code. Not used by the package.


The base year of migration.


Type of migration. Zero means migration is evenly distributed over each time interval. Code 9 means migration is captured at the end of each interval.


Dummy indicating if the country has generalized HIV/AIDS epidemics.


Type of mortality age pattern. Only relevant for countries with the entry “Model life tables”. In such a case, the b_x Lee-Carter parameter is not estimated from historical data. Instead is taken from the dataset MLTbx using a pattern given in the AgeMortalityPattern column.


If AgeMortalityType is equal to “Model life tables”, this value determines which b_x is selected from the MLTbx dataset. It must sorrespond to one of the rownames of MLTbx, e.g. “CD East”, “CD West”, “UN Latin American”.


Method for projecting age-specific mortality rates. It is one of “LC” (modified Lee-Carter, uses function mortcast), “PMD” (pattern mortality decline, uses function copmd), “modPMD” (modified pattern mortality decline, uses function copmd(... use.modpmd = TRUE)), “MLT” (model life tables, uses function mlt), “LogQuad” (log quadratic method, uses function logquad), or “HIVmortmod” (HIV model life tables as implemented in the HIV.LifeTables package which can be installed from the PPgP/HIV.LifeTables GitHub repo).


If the mortality rates are to be projected via a blend of two methods (see mortcast.blend), this column determines the second method. The options are the same as in the column AgeMortProjMethod1.


If one of the AgeMortProjMethodX colums contains the “MLT” method, this column determines the type of the life table (see the argument type in the mlt function).


If the mortality rates are to be projected via a blend of two methods, this column determines the weights in the first and the last year of the projection, respectively. It should be given as an R vector, e.g. “c(1, 0.5)” (see the argument weights in mortcast.blend).


Code determining how the “PMD” method should be adjusted if it's used. 0 means no adjustment, 1 means the argument sexratio.adjust in copmd is set to TRUE, and code 3 means that the argument adjust.sr.if.needed in copmd is set to TRUE.

LatestAgeMortalityPattern, LatestAgeMortalityPattern1

Indicator n for how many latest time periods should be used to compute the a_x Lee-Carter and modPMD parameter. If n is zero, all time periods are used. If n is one, only the latest time period is used. If n is negative, the latest n time periods are excluded. This can have also a form of vector of size 2, where the first element (must be negative) determines how many latest time periods should be excluded, while the second element (must be positive) determines how many latest time periods to include after the exclusion. E.g. “c(-2, 3)” would mean exclude 2 time periods and use the following 3 time points. If the LatestAgeMortalityPattern1 column is present, it should contain values related to an annual simulation (1x1) while the LatestAgeMortalityPattern column relates to a 5x5 simulation.


If LatestAgeMortalityPattern is 1, this column indicates if the a_x should be smoothed.

SmoothDFLatestAgeMortalityPattern, SmoothDFLatestAgeMortalityPattern1

Degree of freedom for smoothing a_x. By default (value 0) a half of the number of age groups is taken. If the SmoothDFLatestAgeMortalityPattern1 column is present, it should contain values related to a 1x1 simulation while the SmoothDFLatestAgeMortalityPattern column relates to a 5x5 simulation.


Type of norm for computing age-specific fertility pattern to which the country belongs to. Currently only “GlobalNorm” is used.

PasfrGlobalNorm, PasfrFarEastAsianNorm, PasfrSouthAsianNorm

Dummies indicating which country to include to compute the specific norms.


There is one record for each country. See Sevcikova et al (2015) on how information from the various columns is used for projections.


Data provided by the United Nations Population Division.


H. Sevcikova, N. Li, V. Kantorova, P. Gerland and A. E. Raftery (2015). Age-Specific Mortality and Fertility Rates for Probabilistic Population Projections. arXiv:1503.05215. https://arxiv.org/abs/1503.05215



bayesPop documentation built on Aug. 10, 2023, 1:10 a.m.