Installation
Install mongolstats and the released mongolmaps from CRAN:
install.packages(c("mongolstats", "mongolmaps"))Your First Analysis: Infant Mortality
This walk-through uses the infant mortality rate (IMR), a key indicator of population health and health system performance. It covers the full workflow: find a table, check its structure, fetch data, calculate rates, and map them.
Step 1: Load Packages
library(mongolstats)
library(mongolmaps)
library(dplyr)
library(ggplot2)
# Set language to English for readable output
# Checked teaching helpers are shipped with the package and can be inspected.
source(system.file("examples", "analysis-helpers.R", package = "mongolstats"), local = TRUE)
source(system.file("examples", "plot-style.R", package = "mongolstats"), local = TRUE)Step 2: Find the Right Table
Search the catalogue by keyword:
mortality_tables <- nso_search("infant mortality")
mortality_tables |>
distinct(tbl_id, tbl_eng_nm)
#> # A tibble: 8 × 2
#> tbl_id tbl_eng_nm
#> <chr> <chr>
#> 1 DT_NSO_2100_014V1 NUMBER OF INFANT MORTALITY, aimags and the Capital and by m…
#> 2 DT_NSO_2100_014V2 NUMBER OF INFANT MORTALITY, by soum and discrict, and by ye…
#> 3 DT_NSO_2100_014V4 INFANT MORTALITY, by sex, by soum, and by year
#> 4 DT_NSO_2100_014V5 INFANT MORTALITY RATE, per 1000 live births, by sex, by sou…
#> 5 DT_NSO_2100_015V1 INFANT MORTALITY RATE, per 1000 live births, aimags and the…
#> 6 DT_NSO_2100_015V2 INFANT MORTALITY, per 1000 live births, by soum, and by year
#> 7 DT_NSO_2100_015V2 INFANT MORTALITY, per 1000 live births, by year
#> 8 DT_NSO_2800_019V1 INFANT MORTALITY RATE PER 1000 LIVE BIRTHSNSO publishes both rates (per 1,000 live births) and
the counts behind them (infant deaths). We start with
the monthly rate table, DT_NSO_2100_015V1, and later use
the counts to compare aimags.
Step 3: Explore Table Metadata
Before fetching data, check which dimensions the table has:
# View table structure
meta <- nso_table_meta("DT_NSO_2100_015V1")
meta
#> # A tibble: 2 × 5
#> dim code is_time n_values codes
#> <chr> <chr> <lgl> <int> <list>
#> 1 Region Бүс FALSE 28 <tibble [28 × 3]>
#> 2 Month Сар TRUE 128 <tibble [128 × 3]>
# Periods oldest first, with their dates. Month labels are "YYYY-MM"; the
# codes count back from the latest month ("0"), so select periods by label.
months <- nso_table_periods("DT_NSO_2100_015V1")
head(months)
#> # A tibble: 6 × 4
#> code label date frequency
#> <chr> <chr> <date> <chr>
#> 1 13 2016-01 2016-01-01 month
#> 2 14 2016-02 2016-02-01 month
#> 3 15 2016-03 2016-03-01 month
#> 4 16 2016-04 2016-04-01 month
#> 5 17 2016-05 2016-05-01 month
#> 6 18 2016-06 2016-06-01 month
range(months$label)
#> [1] "2016-01" "2026-08"Step 4: Fetch Data
Get the national rate for every month NSO publishes:
imr_national <- nso_data(
tbl_id = "DT_NSO_2100_015V1",
selections = list(
Region = "0", # national total
Month = months$label
),
labels = "en" # add English label columns
)
# The Month column holds labels such as "2024-03", which keep their meaning
# when NSO publishes a new month
imr_national |>
arrange(Month) |>
head(10)
#> # A tibble: 10 × 5
#> Region Month value Region_en Month_en
#> <chr> <chr> <dbl> <chr> <chr>
#> 1 0 2016-01 12 Total 2016-01
#> 2 0 2016-02 13 Total 2016-02
#> 3 0 2016-03 14 Total 2016-03
#> 4 0 2016-04 15 Total 2016-04
#> 5 0 2016-05 15 Total 2016-05
#> 6 0 2016-06 14 Total 2016-06
#> 7 0 2016-07 14 Total 2016-07
#> 8 0 2016-08 15 Total 2016-08
#> 9 0 2016-09 12 Total 2016-09
#> 10 0 2016-10 12 Total 2016-10NSO rounds these monthly rates to whole numbers, which is coarse for a measure that sits around 10 to 20 per 1,000.
Step 5: Visualize the Trend
imr_national |>
mutate(date = nso_period_date(Month)) |> # "YYYY-MM" to Date
ggplot(aes(x = date, y = value)) +
geom_line(color = "#183442", linewidth = 1, alpha = 0.3) + # dim raw data so trend stands out
geom_point(color = "#C2623D", size = 2, shape = 21, fill = "white", stroke = 1, alpha = 0.6) +
# Descriptive smooth only: no model-based uncertainty ribbon
geom_smooth(
method = "loess", formula = y ~ x, se = FALSE,
color = "#167D8D", fill = "#167D8D", alpha = 0.2, linewidth = 1.5
) +
scale_x_date(date_breaks = "1 year", date_labels = "%Y") +
scale_y_continuous(limits = c(0, NA)) +
labs(
title = "Infant Mortality Rate in Mongolia (Monthly)",
subtitle = "Deaths under age 1 per 1,000 live births, national",
x = NULL,
y = "IMR (per 1,000)",
caption = "Source: NSO Mongolia, DT_NSO_2100_015V1"
) +
theme(
panel.grid.minor = element_blank(),
panel.grid.major.x = element_blank() # vertical gridlines clutter time series
)The smooth is an exploratory summary of rounded monthly rates. It does not account for different numbers of births or dependence between neighbouring months.
Regional Comparison
To compare aimags over a year, do not average the 12 monthly rates. A mean of monthly rates gives a month with few births the same weight as a busy month, and NSO’s rounding makes small aimags’ monthly rates jump between values. Instead, add up the infant deaths and the live births for the year, then divide:
This conventional same-period vital-statistics IMR is not a linked
birth-cohort probability: infant deaths can include children born in the
preceding year. The counts come from two monthly tables: infant deaths
(DT_NSO_2100_014V1) and live births
(DT_NSO_2100_018V5).
# Expected independent units: national total, 21 aimags and capital code 5.
aimag_codes <- setdiff(mn_aimags()$nso_code, "511")
units <- c("0", aimag_codes, "5")
death_source <- analysis_fetch_counts("DT_NSO_2100_014V1", units)
birth_source <- analysis_fetch_counts("DT_NSO_2100_018V5", units)
# A year must have every calendar month in both tables for all expected units.
# Unresolved observations are retained as missing, never assumed to be zero.
latest_year <- analysis_latest(analysis_common_years(death_source$annual, birth_source$annual))
infant_deaths <- filter(death_source$annual, year == latest_year)
live_births <- filter(birth_source$annual, year == latest_year)
imr_regional <- analysis_pair(
select(infant_deaths, Region, Region_en, deaths = count, missing_deaths = n_missing),
select(live_births, Region, births = count, missing_births = n_missing),
by = "Region"
) |>
mutate(
Region_en = trimws(Region_en),
level = case_when(Region == "0" ~ "National", Region == "5" ~ "Capital", .default = "Aimag")
)
imr_regional <- bind_cols(imr_regional, analysis_poisson(imr_regional$deaths, imr_regional$births)) |>
rename(imr = rate)
national_imr <- imr_regional$imr[imr_regional$Region == "0"]
stopifnot(length(national_imr) == 1L, is.finite(national_imr))
imr_units <- filter(imr_regional, Region != "0")
stopifnot(nrow(imr_units) == 22L)
imr_units |>
transmute(
area = Region_en, deaths, births, imr,
interval = if_else(is.finite(imr), sprintf("%.1f to %.1f", lower, upper), "Insufficient data")
) |>
knitr::kable(
col.names = c("Area", "Infant deaths", "Live births", "IMR per 1,000", "95% CI"),
digits = c(0, 0, 0, 1, 0), format.args = list(big.mark = ",")
)| Area | Infant deaths | Live births | IMR per 1,000 | 95% CI |
|---|---|---|---|---|
| Dornod | 8 | 1,251 | 6.4 | 2.8 to 12.6 |
| Sukhbaatar | 8 | 814 | 9.8 | 4.2 to 19.4 |
| Khentii | 10 | 980 | 10.2 | 4.9 to 18.8 |
| Tuv | 12 | 666 | 18.0 | 9.3 to 31.5 |
| Govisumber | 3 | 287 | 10.5 | 2.2 to 30.5 |
| Selenge | 11 | 991 | 11.1 | 5.5 to 19.9 |
| Dornogovi | 13 | 862 | 15.1 | 8.0 to 25.8 |
| Darkhan-Uul | 18 | 1,800 | 10.0 | 5.9 to 15.8 |
| Umnugovi | 15 | 1,113 | 13.5 | 7.5 to 22.2 |
| Dundgovi | 4 | 514 | 7.8 | 2.1 to 19.9 |
| Orkhon | 15 | 1,936 | 7.7 | 4.3 to 12.8 |
| Uvurkhangai | 28 | 1,609 | 17.4 | 11.6 to 25.2 |
| Bulgan | 7 | 460 | 15.2 | 6.1 to 31.4 |
| Bayankhongor | 8 | 1,268 | 6.3 | 2.7 to 12.4 |
| Arkhangai | 16 | 999 | 16.0 | 9.2 to 26.0 |
| Khuvsgul | 32 | 1,764 | 18.1 | 12.4 to 25.6 |
| Zavkhan | 9 | 830 | 10.8 | 5.0 to 20.6 |
| Govi-Altai | 11 | 813 | 13.5 | 6.8 to 24.2 |
| Bayan-Ulgii | 41 | 2,449 | 16.7 | 12.0 to 22.7 |
| Khovd | 18 | 1,694 | 10.6 | 6.3 to 16.8 |
| Uvs | 28 | 1,366 | 20.5 | 13.6 to 29.6 |
| Ulaanbaatar | 423 | 29,671 | 14.3 | 12.9 to 15.7 |
In 2025, the national recorded IMR was 13.6 per 1,000 live births. 0 areas have insufficient observations for an annual estimate. The source objects retain raw observations, symbols and provenance for inspection.
Visualize Regional Differences
imr_units |>
filter(is.finite(imr)) |>
mutate(Region_en = forcats::fct_reorder(Region_en, imr)) |> # order by rate, not alphabet
ggplot(aes(x = imr, y = Region_en)) +
geom_vline(xintercept = national_imr, linetype = "dashed", color = "grey50") +
geom_errorbar(aes(xmin = lower, xmax = upper), width = 0.3, color = "grey60", orientation = "y") +
geom_point(aes(color = level), size = 2.5) +
scale_color_manual(values = c(Aimag = "#167D8D", Capital = "#C2623D"), name = NULL) +
labs(
title = paste0("Infant Mortality by Aimag, ", latest_year),
subtitle = "Per 1,000 live births, with exact 95% CIs; dashed line = national rate",
x = "Deaths per 1,000 live births",
y = NULL,
caption = "Source: NSO Mongolia, DT_NSO_2100_014V1 and DT_NSO_2100_018V5"
) +
theme(
panel.grid.major.y = element_blank(),
panel.grid.minor = element_blank(),
legend.position = "top"
)The intervals describe Poisson count variation conditional on the recorded counts and birth denominator. They do not include under-registration, revisions or geographic mismatch. Crossing the estimated national line is a descriptive comparison, not an independent two-sample test: the national total includes each aimag. If births or deaths are assigned by place of occurrence rather than residence, referral patterns can distort area comparisons; confirm the definitions of these specific source tables before interpreting differences.
Adding Geographic Context
mongolmaps
provides boundaries that carry NSO codes, so the Region
column joins without matching names.
# mn_join() takes Ulaanbaatar's figures from region code "5" when the
# capital's own code "511" is empty, as it is in this table
imr_map <- imr_units |>
mn_join(by = "Region", level = "aimag")
stopifnot(nrow(imr_map) == 22L, !anyDuplicated(imr_map$nso_code))
mn_map(
imr_map,
fill = imr,
title = paste0("Infant Mortality Rate, ", latest_year)
) +
labs(subtitle = "Infant deaths per 1,000 live births. Data: NSO Mongolia", fill = "IMR\n(per 1,000)") +
guides(fill = guide_colourbar(barwidth = grid::unit(85, "mm"), barheight = grid::unit(3, "mm"))) +
analysis_map_theme()Grey areas have insufficient data. The map shows point estimates only. Read it with the confidence intervals above: in small aimags, a few more or fewer infant deaths can move the colour a long way.
Key Functions Summary
| Function | Purpose | Example |
|---|---|---|
nso_search() |
Find tables by keyword | nso_search("mortality") |
nso_table_meta() |
Get table dimensions | nso_table_meta("DT_NSO_...") |
nso_dim_values() |
List dimension values | nso_dim_values(tbl, "Region") |
nso_table_periods() |
Check time coverage | nso_table_periods(tbl) |
nso_latest_periods() |
Latest periods, for selections | list(Year = nso_latest_periods(tbl, n = 5)) |
nso_data() |
Fetch data | nso_data(tbl, selections, labels) |
nso_period_date() |
Period labels to dates | nso_period_date(data$Month) |
mongolmaps::mn_join() |
Join NSO data to boundaries by code | mn_join(data, by = "Region", level = "aimag") |
mongolmaps::mn_map() |
Quick map of joined data | mn_map(joined, fill = value) |
Best Practices
-
Use codes in scripts, labels for display:
Selections accept codes or labels, but codes such as region codes are
more stable.
labels = "en"adds readable label columns alongside the codes. Periods are the exception: NSO numbers them from the latest ("0"), so mongolstats always returns them as labels, and you should select them by label or withnso_latest_periods(). -
Check metadata first: Use
nso_table_meta()andnso_dim_values()to see dimensions and codes before fetching. -
Separate the levels: Region columns mix the
national total (
"0"), regions ("1"–"4"), the capital ("5"), and aimags. Filter to one level before summing or ranking, or you will count deaths twice. - Build rates from counts: Add up numerators and denominators over the period, then divide. Do not average monthly rates.
- Show uncertainty: Small populations give unstable rates. Report confidence intervals, or pool several years.
-
Join by code: Use
mongolmaps::mn_join()on theRegioncode column rather than matching names.
Common Workflows
Next Steps
- Discover More Data: See the Discovery Guide for advanced search techniques
- Create Maps: Learn spatial analysis in the Mapping Guide
- Reference: Browse all functions in the Reference
Quick Reference: Common Health Tables
| Indicator | Table ID | Measure | Detail |
|---|---|---|---|
| Infant mortality rate | DT_NSO_2100_015V1 | Rate per 1,000 live births | Aimag, month |
| Infant deaths | DT_NSO_2100_014V1 | Count | Aimag, month |
| Live births | DT_NSO_2100_018V5 | Count | Aimag, month |
| Under-5 mortality rate | DT_NSO_2100_030V2 | Rate per 1,000 live births | Aimag, month |
| Maternal deaths | DT_NSO_2100_002V2 | Count | Aimag, month |
| Deaths by leading cause | DT_NSO_2100_027V1 | Count | Soum, year |
| Cancer incidence | DT_NSO_2100_012V1 | Rate per 10,000 population | Cancer type, year |
| Communicable disease cases (incl. tuberculosis) | DT_NSO_2100_035V1 | Count | Disease, month |
| Communicable disease cases | DT_NSO_2100_020V2 | Count | Disease, soum, year |
| Mid-year population | DT_NSO_0300_002V1 | Count | Aimag, year |