The Data Space

Input data shared by every by-policy projection.

The eight input CSVs are read here, once per model, and referenced from Projection as data. Projection is parameterized by point_id, so each Projection[N] is a separate ItemSpace with its own cells cache; if the readers lived there, every model point would re-read every file. Holding them in an unparameterized Space reads each file once no matter how many policies are projected.

Inputs are external files: plain CSVs in the model folder’s parent directory, products/cancer/, rather than data stored inside the model. The model folder therefore holds nothing but formulas – no _data/, no IOSpec, no embedded values – so a diff of the model shows logic changes only. This follows annuallife.TradLife_A; contrast basiclife.BasicTerm_S, which keeps its inputs inside the model through modelx’s IOSpec machinery.

The consequence worth knowing: the model is not portable on its own. Copying the Cancer_KR_S folder without its parent’s CSVs produces a model that reads and then fails on first evaluation.

input_dir() resolves the directory from _model.path.parent at run time, so the model works wherever the repository is checked out. Each table has a filename Reference and a reader Cells:

Reference	Cells	File
model_point_file	model_point_table()	model_point_table.csv
mort_table_file	mort_table()	mort_table.csv
incidence_table_file	incidence_table()	incidence_table.csv
tier_share_file	tier_share_table()	tier_share_table.csv
tier_table_file	tier_table()	tier_table.csv
survival_table_file	survival_table()	survival_table.csv
care_table_file	care_table()	care_table.csv
lapse_table_file	lapse_table()	lapse_table.csv

Every file but model_point_table.csv carries a provenance column, and every cell in it begins with a citation tag. That is not decoration on this product: the eight files sit at three quite different levels of authority and the column is where a reader is told which.

The incidence basis is sourced, and the mortality basis is not

The two decrement files are the opposite of one another, and the contrast is the single most important fact about this model’s inputs.

incidence_table.csv is published data reproduced verbatim. 보험개발원 (the Korea Insurance Development Institute) displays its 장기손해보험 참조순보험요율 in force from 2024-04-01, and that display carries a 「기타피부암 및 갑상선암 이외의 암 발생률」 grid by age and sex [R5] [REG-R61]. Its definition is the insured one – invasive cancer excluding 기타피부암 (C44) and 갑상선암 (C73), classified by 원발부위 – so it already embodies the tier carve-out the 약관 make and the 원발부위 기준 the supervisor imposed from 2011-04-01. Only the two rows above the published age-80 endpoint are [std], and they say so.

mort_table.csv is a [std] construction. Korea’s industry table, the 제10회 경험생명표 applied from 2024-04, is not published in full: 보험개발원 releases the 평균수명 and the 기대여명 and not the rates [REG-R33] [REG-R34]. There is no Korean equivalent of the freely downloadable Japanese 標準生命表, so there is no published rate to anchor on. What is shipped instead is a Makeham q(x) = 1 - exp(-(A + B c^(x+0.5))) whose two free parameters are solved so that the table reproduces the 국가데이터처 생명표’s 2024 기대여명 at ages 40 and 65 exactly – 남 41.9 / 19.5 and 여 47.4 / 23.7 [REG-R38]. It then returns 기대수명 at birth of 80.80 and 86.88 against the published 80.8 and 86.6, which is a check rather than a target and is the only external validation available. Drop a licensed extract in over the same schema – sex, age, mort_rate – and no formula changes.

Four files carry the standardizations, and each says which

tier_share_table.csv splits the sourced base rate into the contract’s tiers. It is [std] and it is where most of the model’s judgement lives: the 특정소액암 and 유사암 shares are anchored on [R1]’s 2023 all-ages crude site rates – 대장 63.8, 유방 58.4, 전립선 44.3, 갑상선 69.3 and 상피내암 74.7 per 100,000 against an excluding-thyroid base of 495.0 – and then graded in age and split by sex, because those all-ages figures mix age distributions that differ violently. The 유사암 ratio is a floor: [R1] does not cover 경계성종양 at all, does not identify 대장점막내암 inside 대장 D010-D012, and does not carry 기타피부암 in its top-ten table.

survival_table.csv is the post-diagnosis basis, and it exists because a cancer contract goes on paying after the diagnosis benefit. Relative survival – 「관찰생존율을 일반인구의 기대생존율로 나누어 구한 값」 [R1] – is a ratio to an expected general-population survival, not a cohort curve and not a transition rate, so it converts into an excess hazard added to the base table rather than a replacement for it. The five select years are calibrated so that the five-year survival ratio equals the published excluding-thyroid figure, 남 65.9% / 여 74.0% [R1]; the 특정소액암 tier is calibrated separately off [R1]’s own site rows; and the 유사암 tier appears in no row of the file at all, because 갑상선 five-year relative survival is 100.2% and lifetime 갑상선 mortality risk 0.1% [R1].

care_table.csv is the weakest file in the model and says so on every row. No Korean source publishes cancer admissions, bed-days, operations or treatment courses per diagnosed patient. The one published utilisation series on the 보험개발원 display is a 질병입원율 for all disease [R5]. The shape is standardized on the clinical ordering the contracts’ own design implies and the level on the 180-day-per-stay cap they carry [S1] [S4] [R3].

lapse_table.csv carries three segments rather than a policy-year grid, because the functional form is prescribed rather than observed: the FSS’s November 2024 ruling makes a 로그-선형 model converging to 0.1% at 완납 the 원칙모형 for 무.저해지 business, with a post-완납 ultimate of 0.8% [REG-R27]. No public Korean lapse or persistency figure for 암보험 exists [R3], so the model implements the prescribed shape and standardizes only its starting level.

tier_table.csv is the benefit ladder itself – 200 / 100 / 60 / 20 per cent of the 보험가입금액 – read directly off the one retrieved 약관 that states every tier as an amount at 보험가입금액 1,000만원 [S3 별표 1]. It also carries each tier’s own 면책기간, which is where the product’s two start dates come from: the invasive tiers attach on the 91st day and the 유사암 tier on the 보험계약일.

Cells Descriptions

input_dir()

The directory holding the input CSVs: the model folder’s parent.

Inputs are external files, not data stored inside the model, so the model folder is pure formulas. The path is resolved at run time from where the model was read, following annuallife.TradLife_A, so the model works from any checkout location.

model_point_table()

The model point table, read from model_point_table.csv.

Indexed by point_id; point_id = 1 is the technical notes’ worked-example anchor cell, male at the 감독규정 기준연령 요건 age of 40 [REG-R9]. premium is an input on this product in a stronger sense than on most: no Korean carrier publishes a rate table for a cancer main contract, the 산출방법서 is a 기초서류 filed with the FSC and not a published document, and the 참조순보험요율 reaches the public only as the 보험가격지수 ratio [REG-R22], so every premium in the table is a [std] modelling value.

This is the one input file exempt from the provenance column, because a model point is a configuration – one policy’s own terms – rather than an assumption.

mort_table()

The [std] all-cause mortality table by sex and age, from mort_table.csv.

A Makeham q(x) = 1 - exp(-(A + B c^(x+0.5))) whose two free parameters reproduce the 국가데이터처 생명표’s 2024 40세 and 65세 기대여명 exactly [REG-R38], not a copy of the 제10회 경험생명표, which 보험개발원 does not publish in full [REG-R33] [REG-R34]. See the Space docstring for why the distinction is load-bearing. Ages 15 to 100, the composite’s issue-age range and its 100세 만기.

incidence_table()

보험개발원 「기타피부암 및 갑상선암 이외의 암 발생률」, from incidence_table.csv.

The 참조순보험요율 display in force from 2024-04-01, by age and sex, on the published ten-year grid [R5] [REG-R61]. Indexed by sex and age; Projection.inc_rate interpolates log-linearly between the grid ages. The two rows above 80 are [std] extrapolations and their provenance says so. These are sourced values reproduced with their attribution, in deliberate contrast to mort_table().

tier_share_table()

The [std] tier decomposition of the base incidence rate, by sex and age.

Three shares per row, from tier_share_table.csv: minor_share – the fraction of the sourced base rate falling in the 특정소액암 tier, its complement being the 일반암 tier; high_share – the 고액암 sub-share, which pays in addition rather than instead; and similar_share – 유사암 incidence expressed as a ratio to the base rate, which is additive to it because the 유사암 tier is outside the base rate’s own definition. Anchored at ages 20, 40, 60 and 80 and interpolated linearly in age.

tier_table()

The benefit ladder and each tier’s own 면책기간, from tier_table.csv.

Four rows: high (고액암, 100% in addition to the general tier, so 200% in total), general (일반암, 100%), minor (특정소액암, 60%) and similar (유사암, 20%), read off the one retrieved 약관 stating every tier as an amount at 보험가입금액 1,000만원 [S3 별표 1]. wait_months is 3 for the invasive tiers and 0 for 유사암, which is the product’s two-start-date structure; waives_premium is 1 for 일반암 and 고액암 alone [S3 제14조제1항] [S1 제9조제1항].

survival_table()

The [std] post-diagnosis excess hazard by sex, tier and select year.

From survival_table.csv, indexed by sex, tier and dur_year 1 to 5 with 6 standing for the ultimate. Calibrated so that exp(-sum of the five annual hazards) reproduces the published 2019-2023 5년 상대생존율 excluding thyroid, 남 65.9% / 여 74.0% [R1], for the general tier, and an [R1]-derived 87.1% / 88.8% for the 특정소액암 tier. The 유사암 tier is absent from the file by design: 갑상선 relative survival is 100.2% and lifetime 갑상선 mortality risk 0.1% [R1], so a 유사암 diagnosis changes no state.

care_table()

The [std] care intensities per diagnosed life, by select year since diagnosis.

From care_table.csv: cancer admissions and mean days per admission, 관혈 and 비관혈 operations, and the annual hazard of the first qualifying 항암약물.방사선 treatment. Indexed by dur_year 1 to 5 with 6 standing for the ultimate. Every figure is a standardization: no Korean source publishes treatment volume per diagnosed cancer patient, and the one published utilisation grid on the 보험개발원 display is a 질병입원율 for all disease rather than for cancer [R5].

lapse_table()

The three-segment [std] lapse basis, from lapse_table.csv.

first_year 4.6%, at_completion 0.1% and post_payment 0.8%. Three segments rather than a policy-year grid because the functional form is prescribed: the FSS’s November 2024 계리가정 ruling makes a 로그-선형 model converging to 0.1% at 완납 the 원칙모형 for 무.저해지 business, with a 0.8% post-완납 ultimate [REG-R27]. Projection.lapse_rate interpolates log-linearly between the first two and steps to the third at 납입완료.