Bundled Actuarial Tables#

Lactuca ships a curated collection of life, disability and exit actuarial tables stored as .ltk binary files. This page explains how to manage the bundled tables and provides a complete reference for every table in the package.

Warning

Data accuracy disclaimer. The actuarial tables bundled with this package are provided as a convenience and are believed to be faithful transcriptions of their original published sources. They are, however, supplied without any warranty — express or implied — as to their accuracy, completeness, or fitness for a particular purpose.

It is the user’s sole responsibility to:

  • verify each table against the official published source before relying on it;

  • assess its suitability for the intended application and jurisdiction;

  • comply with any applicable regulatory requirements.

The author accepts no liability for errors or omissions in the table data, nor for any decision made in reliance on it.

Managing bundled tables#

Tables are bundled with the package but must be written to disk as .ltk files before use. The Tables class in lactuca.tables.data handles all install and discovery operations.

Installing tables#

Note

For generational tables, the cohort parameter (birth year) must be an integer in the range 1900 – current year. Values outside this range raise a ValueError at construction time.

from lactuca.tables.data import Tables

# Install all bundled tables to the default directory (Config.tables_path)
Tables.install()

# Install a single table by its Python variable name
Tables.install("PER2020_Ind_1o")

# Install multiple tables at once (list or tuple)
Tables.install(["PER2020_Ind_1o", "PASEM2020_Rel_1o", "DAV2004R_Agg_2o"])

# Install to a custom directory for this call only (overrides Config.tables_path)
Tables.install(tables_path="/srv/actuarial/tables")

# Overwrite existing .ltk files (default: overwrite=False, skips existing)
Tables.install("PER2020_Ind_1o", overwrite=True)

# The return value depends on the call form:
# - single string → bool (True = installed, False = skipped)
# - list or None  → dict[str, bool] mapping each filename to its outcome
result = Tables.install(["PER2020_Ind_1o", "PASEM2020_Rel_1o"])
print(result)   # {'PER2020_Ind_1o.ltk': True, 'PASEM2020_Rel_1o.ltk': False}
                # False means the file already existed and was not overwritten

Discovering available and missing tables#

from lactuca.tables.data import Tables

# List all installable filenames (does not check disk — returns all known tables)
available = Tables.list_available()
print(available[:3])   # ['PER2020_Col_2o.ltk', 'PER2020_Col_1o.ltk', 'PER2020_Ind_2o.ltk']
                       # order reflects internal registration order; may vary across releases

# List .ltk filenames not yet present in Config.tables_path
missing = Tables.list_missing()
print(missing)         # [] when all tables are installed, or e.g. ['IASS90.ltk']

# Check against a specific directory instead of the configured default
missing_custom = Tables.list_missing(tables_path="/srv/actuarial/tables")

# Install only the missing tables — safe to call repeatedly (idempotent)
result = Tables.install_missing()      # dict[str, bool] — only newly installed tables appear
print(result)          # {'IASS90.ltk': True}  — {} when all tables were already present

Accessing a single table entry#

from lactuca.tables.data import Tables

# Retrieve a TableEntry by Python variable name (returns None if not found)
entry = Tables.get("PASEM2020_Rel_1o")
if entry is not None:
    print(entry.varname)               # 'PASEM2020_Rel_1o'  (also the .ltk file stem)
    print(entry.payload["table_name"]) # 'PASEM 2020 Relativo 1er orden'
    entry.install()                    # install this table individually

The default install directory is actuarial_tables/ at the project root, controlled by Config.tables_path. Each table is saved under its Python variable name; for example, PER2020_Ind_1o is saved as PER2020_Ind_1o.ltk.

Parameter

Type

Default

Description

items

str, list[str], or None

None

Table(s) to install; None installs all

filename

str or None

None

Override the on-disk filename for single-table installs (ignored for batch)

tables_path

str or None

None

Override install directory for this call

overwrite

bool

False

Replace existing .ltk files

strict

bool

True

Raise on validation errors; False emits warnings

See Configuration for how to change the default install path.

Need a table that is not listed here?#

If the table you need is not bundled with the package, you have two options:

Request it. Open a feature request on the Support page describing the table (name, source, jurisdiction, and intended use), and attach the data in a readable format — plain text, CSV, Excel, or any other tabular form. Requests without accompanying data cannot be prioritised. If the table is publicly available and relevant to the actuarial community, it may be added to a future release.

Build it yourself. Any table whose decrement structure fits one of the categories described in Table Taxonomy — aggregate or select-ultimate, static or generational — can be created with TableBuilder and saved as a .ltk file. See Building Custom Table Files (.ltk) for a step-by-step guide.

Table reference#

Column guide:

  • File — Python variable name, also the .ltk file stem (e.g. PER2020_Ind_1o.ltk).

  • Sex — sex coverage: m/f = both sexes, m = male only, f = female only, u = unisex.

  • Country — origin country or publishing organisation.

  • Type — taxonomy code (see Table Taxonomy): combines the class (life / disability / exit) with the temporal–structure cell (agg–static, agg–gen-a, agg–gen-b, agg–gen-c, sel–static, sel–gen-a, sel–gen-b).

  • Order — actuarial loading: 1st = prudential (pricing/reserves); 2nd = best-estimate; — = not applicable.

Life tables#

File

Sex

Country

Type

Order

PER2020_Ind_2o

m/f

Spain (DGSFP)

life · agg–gen-a

2nd

PER2020_Ind_1o

m/f

Spain (DGSFP)

life · agg–gen-a

1st

PER2020_Col_2o

m/f

Spain (DGSFP)

life · agg–gen-a

2nd

PER2020_Col_1o

m/f

Spain (DGSFP)

life · agg–gen-a

1st

PASEM2020_Gen_2o

m/f

Spain (DGSFP)

life · agg–static

2nd

PASEM2020_Dec_2o

m/f

Spain (DGSFP)

life · agg–static

2nd

PASEM2020_Rel_1o

m/f

Spain (DGSFP)

life · agg–static

1st

PASEM2020_NoRel_1o

m/f

Spain (DGSFP)

life · agg–static

1st

PASEM2020_Dec_1o

m/f

Spain (DGSFP)

life · agg–static

1st

PASEM2010

m/f

Spain (DGSFP)

life · agg–static

—

DAV2004R_Agg_2o

m/f

Germany (DAV)

life · agg–gen-b

2nd

DAV2004R_Agg_1o

m/f

Germany (DAV)

life · agg–gen-a

1st

DAV2004R_SelUlt_2o

m/f

Germany (DAV)

life · sel–gen-b

2nd

DAV2004R_SelUlt_1o

m/f

Germany (DAV)

life · sel–gen-a

1st

GAM71

m/f

USA (SOA)

life · agg–static

—

GAM83

m/f

USA (SOA)

life · agg–static

—

GAM94_AA

m/f

USA (SOA)

life · agg–gen-a †

—

AM92_AF92

m/f

UK (CMIB)

life · sel–static ‡‡

—

CB_H_2020

m

Chile (CMF)

life · agg–gen-c

—

MI_H_2020

m

Chile (CMF)

life · agg–gen-c

—

RV_M_2020

f

Chile (CMF)

life · agg–gen-c

—

MI_M_2020

f

Chile (CMF)

life · agg–gen-c

—

B_M_2020

f

Chile (CMF)

life · agg–gen-c

—

† GAM94_AA applies the discrete Scale AA improvement formula (flat age-indexed factors), placing it structurally in the same class as type-a tables.

‡‡ AM92_AF92 is a static select-ultimate table using the CMI/UK Duration-0 convention (start_duration=0): the first select year is duration=0 (unlike most tables where it is duration=1).

Disability tables#

File

Sex

Country

Type

Order

PEAI2007_IAP_Ind

m/f

Spain (DGSFP)

disability · agg–static

—

PEAI2007_IAP_Col

m/f

Spain (DGSFP)

disability · agg–static

—

IASS90

u ‡

Spain (SS)

disability · agg–static

—

SS90TOT

u ‡

Spain (SS)

disability · agg–static

—

SS90ABS

u ‡

Spain (SS)

disability · agg–static

—

‡ sex_independent = True: these tables publish a single rate vector shared by both sexes. Passing sex='m' or sex='f' both return the same rates. The unisex_blend parameter has no effect on these tables.

Development and test tables#

The tables below are synthetic and are intended only for unit testing and development. They are not suitable for production actuarial calculations.

File

Sex

Type

Notes

Dummy_qx0

m/f

life · agg–static

All-zero mortality; verifies annuity = financial annuity

DummyLIFE_1Sm

m

life · agg–static

Static life table, male only

DummyLIFE_1Sf

f

life · agg–static

Static life table, female only

DummyLIFE_LinearGen

m/f

life · agg–gen-a

Synthetic generational life table — linear improvement formula; synthetic, not suitable for production

DummyLIFE_DiscreteGen

m/f

life · agg–gen-a

Synthetic generational life table — discrete improvement formula; synthetic, not suitable for production

DummyLIFE_ProjectedGen

m/f

life · agg–gen-c

Synthetic generational life table — projected improvement formula; synthetic, not suitable for production

DummyLIFE_Select

m/f

life · sel–static

Synthetic select-ultimate life table, select_period=2; synthetic, not suitable for production

DummyLIFE_SelectGen_c

m/f

life · sel–gen-c

Synthetic select-ultimate life table — projected improvement formula, select_period=2, ages 0–100; synthetic, not suitable for production

DummyLIFE_SelectGen_d

m/f

life · sel–gen-d

Synthetic select-ultimate life table — per-duration exponential MI, select_period=2, ages 0–100; synthetic, not suitable for production

DummyLIFE_Unisex

u

life · agg–static

Synthetic sex-independent life table; synthetic, not suitable for production

DummySD2015

m/f

disability · agg–static

Synthetic disability table

DummySD2015Gen

m/f

disability · agg–gen-a

Synthetic generational disability table

DummySD_LinearGen

m/f

disability · agg–gen-a

Synthetic generational disability table — linear improvement formula, ages 20–65; synthetic, not suitable for production

DummySD_DiscreteGen

m/f

disability · agg–gen-a

Synthetic generational disability table — discrete improvement formula, ages 20–65; synthetic, not suitable for production

DummySD_Gen2o

m/f

disability · agg–gen-b

Synthetic generational disability table — year-indexed exponential improvement, ages 20–65; synthetic, not suitable for production

DummySD_ProjectedGen

m/f

disability · agg–gen-c

Synthetic generational disability table — projected improvement formula, ages 20–65; synthetic, not suitable for production

DummySD_Select

m/f

disability · sel–static

Synthetic select-ultimate disability table, ages 20–65, select_period=2; synthetic, not suitable for production

DummySD_SelectGen_a

m/f

disability · sel–gen-a

Synthetic select-ultimate disability table — constant exponential MI, uniform across select durations, select_period=2, ages 20–65; synthetic, not suitable for production

DummySD_SelectGen_b

m/f

disability · sel–gen-b

Synthetic select-ultimate disability table — year-indexed MI, uniform across select durations, select_period=2, ages 20–65; synthetic, not suitable for production

DummySD_SelectGen_c

m/f

disability · sel–gen-c

Synthetic select-ultimate disability table — projected improvement formula, select_period=2, ages 20–65; synthetic, not suitable for production

DummySD_SelectGen_d

m/f

disability · sel–gen-d

Synthetic select-ultimate disability table — per-duration exponential MI, select_period=2, ages 20–65; synthetic, not suitable for production

DummySD_Unisex

u

disability · agg–static

Synthetic sex-independent disability table, ages 20–65; synthetic, not suitable for production

DummyEXIT

m/f

exit · agg–static

Synthetic exit/withdrawal table

DummyEXIT_Gen

m/f

exit · agg–gen-a

Synthetic generational exit table

DummyEXIT_LinearGen

m/f

exit · agg–gen-a

Synthetic generational exit table — linear improvement formula; synthetic, not suitable for production

DummyEXIT_DiscreteGen

m/f

exit · agg–gen-a

Synthetic generational exit table — discrete improvement formula; synthetic, not suitable for production

DummyEXIT_Gen2o

m/f

exit · agg–gen-b

Synthetic generational exit table — year-indexed exponential improvement, ages 0–100; synthetic, not suitable for production

DummyEXIT_ProjectedGen

m/f

exit · agg–gen-c

Synthetic generational exit table — projected improvement formula; synthetic, not suitable for production

DummyEXIT_Select

m/f

exit · sel–static

Synthetic select-ultimate exit table, select_period=2; synthetic, not suitable for production

DummyEXIT_SelectGen_a

m/f

exit · sel–gen-a

Synthetic select-ultimate exit table — constant exponential MI, uniform across select durations, select_period=2, ages 0–100; synthetic, not suitable for production

DummyEXIT_SelectGen_b

m/f

exit · sel–gen-b

Synthetic select-ultimate exit table — year-indexed MI, uniform across select durations, select_period=2, ages 0–100; synthetic, not suitable for production

DummyEXIT_SelectGen_c

m/f

exit · sel–gen-c

Synthetic select-ultimate exit table — projected improvement formula, select_period=2, ages 0–100; synthetic, not suitable for production

DummyEXIT_SelectGen_d

m/f

exit · sel–gen-d

Synthetic select-ultimate exit table — per-duration exponential MI, select_period=2, ages 0–100; synthetic, not suitable for production

DummyEXIT_Unisex

u

exit · agg–static

Synthetic sex-independent exit table; synthetic, not suitable for production

See also#