The Data Space

Input data shared by every by-policy projection.

The seven 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/child/, 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 Child_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_file	incidence_table()	incidence_table.csv
basis_file	basis_table()	basis_table.csv
neonatal_file	neonatal_table()	neonatal_table.csv
lapse_table_file	lapse_table()	lapse_table.csv
av_table_file	av_table()	av_table.csv

The data position for this product is the worst in the library

No Korean mortality table in this library is a published table. 경험생명표 (gyeongheom saengmyeongpyo, the industry experience life table, 제10회 applied from 2024-04) is produced by 보험개발원 and is not released in full: what is published is the summary — the 평균수명 and the 기대여명 — and not the rates [REG-R33] [REG-R34]. mort_table.csv is therefore a [std] construction, log-linear between anchor ages shaped on the 통계청 완전생명표 age pattern [REG-R38] [REG-R39], carrying the infant peak, the childhood trough at about age 10 and the adolescent turn that a child policy is exposed to for its first two decades. It is used for the insured child and for the 계약자, whose death is the second of the two waiver triggers.

The morbidity position is worse, though not for the reason a first reading suggests. Each carrier’s own 적용위험률 lives in the 산출방법서, an undisclosed 기초서류 [REG-R2], and the bureau’s 참조순보험요율 is filed under 보험업법 제176조제4항 with no general obligation to publish [REG-R4]. But 제176조제9항 lets the bureau publish 순보험요율 산출 자료 where policyholder protection requires it, and for 장기손해보험 it does: a dated display carrying 「기타피부암 및 갑상선암 이외의 암 발생률」, 질병입원율 and 후유장해 by age and sex, whose age grid reaches 연령 0 and 10 [REG-R61]. This product’s research pass never opened it — the cancer and indemnity_medical passes did — and nothing on child incidence was retrieved from 보험개발원, 국가암정보센터 or 통계청 in this pass. Every rate in incidence_table.csv is therefore a [std] construction whose provenance cell names the authority its shape rests on: the 국가암등록통계 연령별 발생률 [REG-R40], the two grids of the 참조순보험요율 display [REG-R61], and the 국민건강보험 진료비 실태조사 [REG-R41]. It is [std] because it was never reconciled to that display, not because no display exists, and a later pass should re-base the rows off it — noting that a 참조순보험요율 is a net premium rate, not a best estimate [REG-R61]. The one exception is the row that anchors the whole file: 일반상해 후유장해 발생률(3~100%), 기본계약, 5세, 상해 1급 — 남자 0.0001823, 여자 0.0001163 [S1], the only observation of a Korean child morbidity rate anywhere in the research, and the point the basic contract’s decrement is calibrated on.

basis_table.csv carries the scalar [std] parameters that turn an incidence into a cost — the mean 장해지급률, the surgery rates given a diagnosis, the liability severity, the per-stay cap factor, the 50%-severity share that fires the waiver, the 계약자’s disability-to-mortality ratio, the 누수사고 share of the liability cost, the foetal-loss rate and the broad-definition factor. neonatal_table.csv carries the 태아 module: nine limbs, each with a frequency, an amount (fixed or a ratio to the module’s 가입금액), an expected number of units and a timing — birth for the 태아보장기간 limbs paid at birth, block for the 1년만기 신생아 block. Two of them are day-capped rather than amount-capped, which is why the module’s cost is a length-of-stay question.

lapse_table.csv carries three bases: the 2024 계리가정 guideline’s 원칙모형 [REG-R27], the 적용해지율 one carrier actually discloses [S1], and a level comparison vector. av_table.csv carries the published 표준형 환급률 progression of a current 상품요약서 [S2] as a build curve indexed by duration in years, and a taper curve indexed by the fraction of the term run off, calibrated to reproduce the published 16.0% at 95 years and 0.0% at 만기 — so that one shipped progression serves every 보험기간.

To swap in a company basis, replace the files with same-schema ones, or point the filename References at different names, then clear the cache. No formula changes.

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.

model_point_table()

The model point table, read from model_point_table.csv.

Indexed by point_id. point_id = 1 is the anchor cell of the technical notes’ worked example — a 태아가입 contract, priced male, 계약나이 0 at the 계약일, birth at policy month 5, 100세만기 20년납 월납, 표준형, with both premium waivers on and the 계약자 male 만 33 — and the other nine exercise the sexes, the issue-age and term envelopes, the three surrender-value forms, the optional modules and the switches.

mort_table()

Annual mortality by sex and 만나이, from mort_table.csv.

A [std] construction, log-linear between anchor ages shaped on the Korean 완전생명표 age pattern [REG-R38] [REG-R39]; the 제10회 경험생명표 is published only as summary statistics [REG-R33] [REG-R34]. It serves two lives: the insured child, whose mortality begins at birth and not at the 계약일, and the 계약자, whose death is one of the two triggers of the premium waiver. The largest age present is 120, where the rate is 1.

incidence_table()

Annual incidence by cause, sex and pivot 만나이, from incidence_table.csv.

Eleven causes — disability, disease_disab, cancer, minor_cancer, cerebral, cardiac, fracture, burn, hosp_acc, hosp_dis and liability — at fourteen pivot ages from 0 to 100, interpolated log-linearly in inc_rate_at(). The two hosp_* causes are expected days per policy year rather than event frequencies, before the 1~180일 per-stay cap; every other cause is an annual event frequency.

Every rate here is a [std] construction and the provenance column says which authority its shape rests on, save the two rows that are not: 일반상해 후유장해 발생률(3~100%) at 5세, 상해 1급 — 남자 0.0001823, 여자 0.0001163 [S1] — the only published Korean child morbidity rate in the whole research file.

basis_table()

The scalar [std] parameters of the benefit basis, from basis_table.csv.

Thirteen values, each with its own provenance: the mean 장해지급률 on an accident and on a disease impairment, the three surgery rates given a diagnosis, the liability severity, the per-stay cap factor on the hospital-cash limbs, the share of 3~100% impairments reaching the 50% waiver threshold, the 계약자’s disability-to-mortality ratio, the 누수사고 share of the liability cost, the foetal-loss rate, the broad-definition factor on the two adult-disease limbs, and the 순보험료 share used in the 표준해약공제액 computation.

neonatal_table()

The 태아 module’s nine limbs, from neonatal_table.csv.

Each row carries a timing — birth for a 태아보장기간 limb paid at birth, block for a limb of the 1년만기 신생아 block — a frequency per birth, a fixed amount, an amount_ratio to the module’s own 가입금액, and the expected number of units. The incubator and perinatal-cash limbs are day-capped rather than amount-capped, so their units are expected paid days after the contractual deduction and inside the cap, which is why the module’s cost is a length-of-stay question rather than an amount question.

lapse_table()

The three lapse bases, from lapse_table.csv.

Each basis carries a first-year rate, the rate reached at 납입완료 and the ultimate rate afterwards; the durational shape between the first two is applied in lapse_rate(). loglinear is the 2024 계리가정 guideline’s 원칙모형 [REG-R27] [R11], disclosed the 적용해지율 one carrier publishes for its suppressed forms [S1], and flat a level comparison vector.

av_table()

The 해약환급금 progression, read from av_table.csv.

Two curves in one file, keyed by the curve column. build is the published 표준형 환급률 grid of a current 상품요약서 — 0.0% at 1 year, 45.6% at 3, 62.5% at 5, 73.7% at 10, 78.3% at 15, 82.6% at 20, 101.2% at 30, 122.5% at 40, 144.1% at 50 and 158.9% at 60 [S2] — indexed by duration in years and held flat beyond 60. taper is the terminal collapse, indexed by the fraction of the 보험기간 run off and calibrated so that a 100세만기 contract reproduces the published 16.0% at 95 years and pays nothing at 만기. Splitting the progression in two is what lets one shipped grid serve every 보험기간 without re-basing the published figures.