FA-61181 / Bond day-count conventions / Open access

Step-up coupon accrued interest: the settlement day itself accrues · case 01

Accrued interest is one day of coupon too high.

Verified by executionVariant 1 · 8 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The accrual loop runs while x <= settle.

VERIFIED REPAIR

Accrue days in [prev, settle).

Unsuccessful approach: Stopping the loop only at the next coupon date still includes the settlement day.

Case contract

Inputs prev and next coupon dates, settlement in [prev, next], a list of [date, rate] steps (unordered; a step applies on and after its date), the base rate before any step, and frequency. Each accrued day d in [prev, settle) earns the rate in force on d. Accrued = 100/freq * sum(rates)/days(prev, next), rounded to 6.

Why this case matters

Bond accrual and pricing systems depend on exact day-count arithmetic; a single-day error changes settlement cash.

1 / The failure

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json
import datetime
from fractions import Fraction
N = 1
observations = []
def solve(prev, nxt, settle, steps, base_rate, freq):
    P = datetime.date(*prev)
    Q = datetime.date(*nxt)
    S = datetime.date(*settle)
    period = (Q - P).days
    sched = sorted((datetime.date(*s[0]), s[1]) for s in steps)
    def rate_on(x):
        r = base_rate
        for when, v in sched:
            if when <= x:
                r = v
        return r
    total = Fraction(0)
    x = P
    while x <= S:
        total += Fraction(str(rate_on(x)))
        x += datetime.timedelta(days=1)
    return round(float(total * 100 / freq / period), 6)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['regression accrual end inclusivity 1', [[2019, 12, 1], [2020, 6, 1], [2020, 5, 9], [], 0.025, 2], 1.092896], ['regression accrual end inclusivity 2', [[2021, 3, 31], [2021, 6, 30], [2021, 6, 9], [], 0.02, 4], 0.384615], ['partial repair probe 1', [[2015, 8, 6], [2016, 2, 6], [2016, 1, 13], [[[2015, 12, 22], 0.05], [[2015, 10, 22], 0.05]], 0.02, 2], 1.546196], ['partial repair probe 2', [[2040, 8, 28], [2041, 8, 28], [2041, 1, 21], [], 0.02, 1], 0.8], ['normal control 1', [[2037, 5, 31], [2038, 5, 31], [2037, 10, 11], [[[2037, 8, 17], 0.04], [[2037, 6, 3], 0.05], [[2037, 5, 6], 0.03]], 0.025, 1], 1.654795], ['normal control 2', [[2022, 10, 4], [2023, 1, 4], [2022, 12, 21], [[[2022, 11, 14], 0.06]], 0.025, 4], 0.881793], ['normal control 3', [[2039, 6, 4], [2039, 12, 4], [2039, 8, 31], [[[2039, 6, 27], 0.04], [[2039, 11, 1], 0.035], [[2039, 8, 29], 0.035], [[2039, 7, 21], 0.05]], 0.02, 2], 0.939891], ['normal control 4', [[2013, 10, 30], [2014, 4, 30], [2014, 1, 30], [[[2014, 3, 23], 0.06], [[2014, 3, 27], 0.07]], 0.02, 2], 0.505495]], [['regression accrual end inclusivity 1', [[2030, 9, 23], [2031, 9, 23], [2030, 11, 20], [[[2030, 12, 5], 0.04], [[2030, 11, 20], 0.08]], 0.025, 1], 0.39726], ['regression accrual end inclusivity 2', [[2017, 3, 21], [2017, 6, 21], [2017, 3, 31], [[[2017, 4, 6], 0.05], [[2017, 6, 17], 0.05]], 0.02, 4], 0.054348], ['partial repair probe 1', [[2020, 5, 30], [2020, 11, 30], [2020, 6, 18], [[[2020, 9, 12], 0.04], [[2020, 5, 19], 0.035]], 0.025, 2], 0.180707], ['partial repair probe 2', [[2040, 8, 27], [2041, 2, 27], [2040, 10, 20], [[[2040, 11, 9], 0.035], [[2040, 7, 31], 0.05], [[2040, 9, 25], 0.05]], 0.02, 2], 0.733696], ['normal control 1', [[2033, 9, 30], [2033, 12, 30], [2033, 10, 8], [], 0.025, 4], 0.054945], ['normal control 2', [[2028, 5, 31], [2029, 5, 31], [2029, 4, 16], [[[2028, 8, 18], 0.06]], 0.02, 1], 4.394521], ['normal control 3', [[2031, 6, 30], [2032, 6, 30], [2032, 6, 30], [[[2031, 10, 19], 0.04], [[2032, 2, 17], 0.035], [[2031, 6, 12], 0.06], [[2032, 1, 13], 0.07]], 0.02, 1], 4.710383], ['normal control 4', [[2020, 3, 1], [2020, 6, 1], [2020, 4, 29], [[[2020, 2, 12], 0.05], [[2020, 4, 6], 0.06], [[2020, 2, 6], 0.06], [[2020, 2, 28], 0.04]], 0.02, 4], 0.766304]], [['regression accrual end inclusivity 1', [[2023, 5, 7], [2024, 5, 7], [2023, 12, 4], [[[2023, 8, 29], 0.05], [[2023, 9, 12], 0.07], [[2023, 6, 15], 0.04], [[2023, 7, 9], 0.035]], 0.025, 1], 2.795082], ['regression accrual end inclusivity 2', [[2025, 8, 31], [2026, 8, 31], [2026, 5, 4], [[[2025, 10, 31], 0.06], [[2025, 12, 24], 0.06], [[2026, 5, 29], 0.035], [[2026, 3, 23], 0.06]], 0.02, 1], 3.375342], ['partial repair probe 1', [[2021, 3, 27], [2022, 3, 27], [2021, 5, 24], [[[2021, 12, 2], 0.07]], 0.025, 1], 0.39726], ['partial repair probe 2', [[2028, 9, 30], [2029, 9, 30], [2029, 8, 29], [[[2029, 8, 9], 0.03], [[2028, 11, 8], 0.05], [[2028, 11, 20], 0.04], [[2029, 8, 29], 0.08]], 0.025, 1], 3.467123], ['normal control 1', [[2022, 2, 28], [2023, 2, 28], [2022, 6, 8], [[[2022, 11, 1], 0.06]], 0.025, 1], 0.684932], ['normal control 2', [[2040, 5, 31], [2040, 11, 30], [2040, 6, 26], [[[2040, 8, 23], 0.03], [[2040, 7, 18], 0.03]], 0.025, 2], 0.177596], ['normal control 3', [[2013, 3, 24], [2014, 3, 24], [2014, 2, 4], [[[2013, 9, 17], 0.06], [[2013, 6, 18], 0.035]], 0.025, 1], 3.763014], ['normal control 4', [[2039, 6, 15], [2040, 6, 15], [2040, 1, 8], [[[2039, 9, 3], 0.03], [[2039, 10, 30], 0.04], [[2040, 5, 12], 0.07]], 0.02, 1], 1.669399]], [['regression accrual end inclusivity 1', [[2036, 9, 12], [2037, 3, 12], [2036, 12, 23], [[[2036, 11, 29], 0.03], [[2036, 10, 31], 0.04]], 0.02, 2], 0.790055], ['regression accrual end inclusivity 2', [[2021, 11, 19], [2022, 5, 19], [2022, 4, 30], [[[2021, 12, 23], 0.04], [[2022, 1, 3], 0.035], [[2022, 1, 21], 0.04], [[2022, 2, 8], 0.07], [[2022, 4, 30], 0.08]], 0.025, 2], 2.29558], ['partial repair probe 1', [[2032, 11, 24], [2033, 11, 24], [2033, 2, 3], [], 0.025, 1], 0.486301], ['partial repair probe 2', [[2021, 11, 7], [2022, 2, 7], [2021, 12, 17], [[[2021, 11, 28], 0.05], [[2022, 2, 12], 0.05], [[2022, 1, 19], 0.03]], 0.025, 4], 0.400815], ['normal control 1', [[2022, 1, 27], [2023, 1, 27], [2022, 4, 18], [[[2022, 5, 16], 0.03], [[2022, 2, 7], 0.06], [[2022, 11, 24], 0.03], [[2022, 4, 20], 0.05]], 0.025, 1], 1.226027], ['normal control 2', [[2028, 9, 1], [2029, 9, 1], [2028, 10, 7], [], 0.02, 1], 0.19726], ['normal control 3', [[2037, 2, 4], [2037, 5, 4], [2037, 4, 22], [[[2037, 4, 8], 0.035]], 0.02, 4], 0.491573], ['normal control 4', [[2015, 10, 1], [2016, 4, 1], [2015, 10, 24], [[[2016, 3, 16], 0.06]], 0.025, 2], 0.157104]], [['regression accrual end inclusivity 1', [[2027, 7, 31], [2028, 7, 31], [2028, 2, 28], [], 0.025, 1], 1.448087], ['regression accrual end inclusivity 2', [[2019, 11, 18], [2020, 2, 18], [2020, 2, 2], [[[2019, 11, 12], 0.06], [[2020, 1, 23], 0.06], [[2020, 1, 23], 0.03]], 0.02, 4], 1.23913], ['partial repair probe 1', [[2038, 4, 20], [2038, 10, 20], [2038, 6, 3], [[[2038, 7, 30], 0.03], [[2038, 10, 2], 0.07], [[2038, 6, 3], 0.08]], 0.02, 2], 0.240437], ['partial repair probe 2', [[2026, 3, 2], [2026, 6, 2], [2026, 5, 3], [[[2026, 4, 24], 0.03], [[2026, 3, 28], 0.07]], 0.02, 4], 0.728261], ['normal control 1', [[2014, 6, 27], [2015, 6, 27], [2015, 1, 25], [], 0.02, 1], 1.161644], ['normal control 2', [[2023, 5, 11], [2023, 11, 11], [2023, 8, 30], [[[2023, 9, 9], 0.03], [[2023, 5, 9], 0.035], [[2023, 8, 30], 0.08]], 0.02, 2], 1.055707], ['normal control 3', [[2010, 1, 30], [2011, 1, 30], [2010, 8, 26], [[[2010, 9, 1], 0.05], [[2010, 2, 10], 0.035], [[2010, 8, 26], 0.08]], 0.025, 1], 1.964384], ['normal control 4', [[2036, 10, 30], [2037, 4, 30], [2037, 2, 1], [[[2036, 10, 9], 0.07]], 0.025, 2], 1.807692]]]
for label, args, expected in fixtures[N-1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
regression accrual end inclusivity 1	`1.099727`	`1.092896`	Failed
regression accrual end inclusivity 2	`0.39011`	`0.384615`	Failed
partial repair probe 1	`1.559783`	`1.546196`	Failed
partial repair probe 2	`0.805479`	`0.8`	Failed
normal control 1	`1.665753`	`1.654795`	Failed
normal control 2	`0.898098`	`0.881793`	Failed
normal control 3	`0.949454`	`0.939891`	Failed
normal control 4	`0.510989`	`0.505495`	Failed

SHA-256 / 6eff81a2a07978831fcf664d780e7ae04f75ecb330e9f772a5c15ba4eaa4c0a4

2 / The unsuccessful fix

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json
import datetime
from fractions import Fraction
N = 1
observations = []
def solve(prev, nxt, settle, steps, base_rate, freq):
    P = datetime.date(*prev)
    Q = datetime.date(*nxt)
    S = datetime.date(*settle)
    period = (Q - P).days
    sched = sorted((datetime.date(*s[0]), s[1]) for s in steps)
    def rate_on(x):
        r = base_rate
        for when, v in sched:
            if when <= x:
                r = v
        return r
    total = Fraction(0)
    x = P
    while x <= S and x < Q:
        total += Fraction(str(rate_on(x)))
        x += datetime.timedelta(days=1)
    return round(float(total * 100 / freq / period), 6)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['regression accrual end inclusivity 1', [[2019, 12, 1], [2020, 6, 1], [2020, 5, 9], [], 0.025, 2], 1.092896], ['regression accrual end inclusivity 2', [[2021, 3, 31], [2021, 6, 30], [2021, 6, 9], [], 0.02, 4], 0.384615], ['partial repair probe 1', [[2015, 8, 6], [2016, 2, 6], [2016, 1, 13], [[[2015, 12, 22], 0.05], [[2015, 10, 22], 0.05]], 0.02, 2], 1.546196], ['partial repair probe 2', [[2040, 8, 28], [2041, 8, 28], [2041, 1, 21], [], 0.02, 1], 0.8], ['normal control 1', [[2037, 5, 31], [2038, 5, 31], [2037, 10, 11], [[[2037, 8, 17], 0.04], [[2037, 6, 3], 0.05], [[2037, 5, 6], 0.03]], 0.025, 1], 1.654795], ['normal control 2', [[2022, 10, 4], [2023, 1, 4], [2022, 12, 21], [[[2022, 11, 14], 0.06]], 0.025, 4], 0.881793], ['normal control 3', [[2039, 6, 4], [2039, 12, 4], [2039, 8, 31], [[[2039, 6, 27], 0.04], [[2039, 11, 1], 0.035], [[2039, 8, 29], 0.035], [[2039, 7, 21], 0.05]], 0.02, 2], 0.939891], ['normal control 4', [[2013, 10, 30], [2014, 4, 30], [2014, 1, 30], [[[2014, 3, 23], 0.06], [[2014, 3, 27], 0.07]], 0.02, 2], 0.505495]], [['regression accrual end inclusivity 1', [[2030, 9, 23], [2031, 9, 23], [2030, 11, 20], [[[2030, 12, 5], 0.04], [[2030, 11, 20], 0.08]], 0.025, 1], 0.39726], ['regression accrual end inclusivity 2', [[2017, 3, 21], [2017, 6, 21], [2017, 3, 31], [[[2017, 4, 6], 0.05], [[2017, 6, 17], 0.05]], 0.02, 4], 0.054348], ['partial repair probe 1', [[2020, 5, 30], [2020, 11, 30], [2020, 6, 18], [[[2020, 9, 12], 0.04], [[2020, 5, 19], 0.035]], 0.025, 2], 0.180707], ['partial repair probe 2', [[2040, 8, 27], [2041, 2, 27], [2040, 10, 20], [[[2040, 11, 9], 0.035], [[2040, 7, 31], 0.05], [[2040, 9, 25], 0.05]], 0.02, 2], 0.733696], ['normal control 1', [[2033, 9, 30], [2033, 12, 30], [2033, 10, 8], [], 0.025, 4], 0.054945], ['normal control 2', [[2028, 5, 31], [2029, 5, 31], [2029, 4, 16], [[[2028, 8, 18], 0.06]], 0.02, 1], 4.394521], ['normal control 3', [[2031, 6, 30], [2032, 6, 30], [2032, 6, 30], [[[2031, 10, 19], 0.04], [[2032, 2, 17], 0.035], [[2031, 6, 12], 0.06], [[2032, 1, 13], 0.07]], 0.02, 1], 4.710383], ['normal control 4', [[2020, 3, 1], [2020, 6, 1], [2020, 4, 29], [[[2020, 2, 12], 0.05], [[2020, 4, 6], 0.06], [[2020, 2, 6], 0.06], [[2020, 2, 28], 0.04]], 0.02, 4], 0.766304]], [['regression accrual end inclusivity 1', [[2023, 5, 7], [2024, 5, 7], [2023, 12, 4], [[[2023, 8, 29], 0.05], [[2023, 9, 12], 0.07], [[2023, 6, 15], 0.04], [[2023, 7, 9], 0.035]], 0.025, 1], 2.795082], ['regression accrual end inclusivity 2', [[2025, 8, 31], [2026, 8, 31], [2026, 5, 4], [[[2025, 10, 31], 0.06], [[2025, 12, 24], 0.06], [[2026, 5, 29], 0.035], [[2026, 3, 23], 0.06]], 0.02, 1], 3.375342], ['partial repair probe 1', [[2021, 3, 27], [2022, 3, 27], [2021, 5, 24], [[[2021, 12, 2], 0.07]], 0.025, 1], 0.39726], ['partial repair probe 2', [[2028, 9, 30], [2029, 9, 30], [2029, 8, 29], [[[2029, 8, 9], 0.03], [[2028, 11, 8], 0.05], [[2028, 11, 20], 0.04], [[2029, 8, 29], 0.08]], 0.025, 1], 3.467123], ['normal control 1', [[2022, 2, 28], [2023, 2, 28], [2022, 6, 8], [[[2022, 11, 1], 0.06]], 0.025, 1], 0.684932], ['normal control 2', [[2040, 5, 31], [2040, 11, 30], [2040, 6, 26], [[[2040, 8, 23], 0.03], [[2040, 7, 18], 0.03]], 0.025, 2], 0.177596], ['normal control 3', [[2013, 3, 24], [2014, 3, 24], [2014, 2, 4], [[[2013, 9, 17], 0.06], [[2013, 6, 18], 0.035]], 0.025, 1], 3.763014], ['normal control 4', [[2039, 6, 15], [2040, 6, 15], [2040, 1, 8], [[[2039, 9, 3], 0.03], [[2039, 10, 30], 0.04], [[2040, 5, 12], 0.07]], 0.02, 1], 1.669399]], [['regression accrual end inclusivity 1', [[2036, 9, 12], [2037, 3, 12], [2036, 12, 23], [[[2036, 11, 29], 0.03], [[2036, 10, 31], 0.04]], 0.02, 2], 0.790055], ['regression accrual end inclusivity 2', [[2021, 11, 19], [2022, 5, 19], [2022, 4, 30], [[[2021, 12, 23], 0.04], [[2022, 1, 3], 0.035], [[2022, 1, 21], 0.04], [[2022, 2, 8], 0.07], [[2022, 4, 30], 0.08]], 0.025, 2], 2.29558], ['partial repair probe 1', [[2032, 11, 24], [2033, 11, 24], [2033, 2, 3], [], 0.025, 1], 0.486301], ['partial repair probe 2', [[2021, 11, 7], [2022, 2, 7], [2021, 12, 17], [[[2021, 11, 28], 0.05], [[2022, 2, 12], 0.05], [[2022, 1, 19], 0.03]], 0.025, 4], 0.400815], ['normal control 1', [[2022, 1, 27], [2023, 1, 27], [2022, 4, 18], [[[2022, 5, 16], 0.03], [[2022, 2, 7], 0.06], [[2022, 11, 24], 0.03], [[2022, 4, 20], 0.05]], 0.025, 1], 1.226027], ['normal control 2', [[2028, 9, 1], [2029, 9, 1], [2028, 10, 7], [], 0.02, 1], 0.19726], ['normal control 3', [[2037, 2, 4], [2037, 5, 4], [2037, 4, 22], [[[2037, 4, 8], 0.035]], 0.02, 4], 0.491573], ['normal control 4', [[2015, 10, 1], [2016, 4, 1], [2015, 10, 24], [[[2016, 3, 16], 0.06]], 0.025, 2], 0.157104]], [['regression accrual end inclusivity 1', [[2027, 7, 31], [2028, 7, 31], [2028, 2, 28], [], 0.025, 1], 1.448087], ['regression accrual end inclusivity 2', [[2019, 11, 18], [2020, 2, 18], [2020, 2, 2], [[[2019, 11, 12], 0.06], [[2020, 1, 23], 0.06], [[2020, 1, 23], 0.03]], 0.02, 4], 1.23913], ['partial repair probe 1', [[2038, 4, 20], [2038, 10, 20], [2038, 6, 3], [[[2038, 7, 30], 0.03], [[2038, 10, 2], 0.07], [[2038, 6, 3], 0.08]], 0.02, 2], 0.240437], ['partial repair probe 2', [[2026, 3, 2], [2026, 6, 2], [2026, 5, 3], [[[2026, 4, 24], 0.03], [[2026, 3, 28], 0.07]], 0.02, 4], 0.728261], ['normal control 1', [[2014, 6, 27], [2015, 6, 27], [2015, 1, 25], [], 0.02, 1], 1.161644], ['normal control 2', [[2023, 5, 11], [2023, 11, 11], [2023, 8, 30], [[[2023, 9, 9], 0.03], [[2023, 5, 9], 0.035], [[2023, 8, 30], 0.08]], 0.02, 2], 1.055707], ['normal control 3', [[2010, 1, 30], [2011, 1, 30], [2010, 8, 26], [[[2010, 9, 1], 0.05], [[2010, 2, 10], 0.035], [[2010, 8, 26], 0.08]], 0.025, 1], 1.964384], ['normal control 4', [[2036, 10, 30], [2037, 4, 30], [2037, 2, 1], [[[2036, 10, 9], 0.07]], 0.025, 2], 1.807692]]]
for label, args, expected in fixtures[N-1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
regression accrual end inclusivity 1	`1.099727`	`1.092896`	Failed
regression accrual end inclusivity 2	`0.39011`	`0.384615`	Failed
partial repair probe 1	`1.559783`	`1.546196`	Failed
partial repair probe 2	`0.805479`	`0.8`	Failed
normal control 1	`1.665753`	`1.654795`	Failed
normal control 2	`0.898098`	`0.881793`	Failed
normal control 3	`0.949454`	`0.939891`	Failed
normal control 4	`0.510989`	`0.505495`	Failed

SHA-256 / ca6b80bf95f0fff80bade9df05eb02300ffe53645603a4dda84edd8045c5cd08

3 / The verified repair

Exit 0

"""Failure Map reference implementation. Python standard library only."""
import json
import datetime
from fractions import Fraction
N = 1
observations = []
def solve(prev, nxt, settle, steps, base_rate, freq):
    P = datetime.date(*prev)
    Q = datetime.date(*nxt)
    S = datetime.date(*settle)
    period = (Q - P).days
    sched = sorted((datetime.date(*s[0]), s[1]) for s in steps)
    def rate_on(x):
        r = base_rate
        for when, v in sched:
            if when <= x:
                r = v
        return r
    total = Fraction(0)
    x = P
    while x < S:
        total += Fraction(str(rate_on(x)))
        x += datetime.timedelta(days=1)
    return round(float(total * 100 / freq / period), 6)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['regression accrual end inclusivity 1', [[2019, 12, 1], [2020, 6, 1], [2020, 5, 9], [], 0.025, 2], 1.092896], ['regression accrual end inclusivity 2', [[2021, 3, 31], [2021, 6, 30], [2021, 6, 9], [], 0.02, 4], 0.384615], ['partial repair probe 1', [[2015, 8, 6], [2016, 2, 6], [2016, 1, 13], [[[2015, 12, 22], 0.05], [[2015, 10, 22], 0.05]], 0.02, 2], 1.546196], ['partial repair probe 2', [[2040, 8, 28], [2041, 8, 28], [2041, 1, 21], [], 0.02, 1], 0.8], ['normal control 1', [[2037, 5, 31], [2038, 5, 31], [2037, 10, 11], [[[2037, 8, 17], 0.04], [[2037, 6, 3], 0.05], [[2037, 5, 6], 0.03]], 0.025, 1], 1.654795], ['normal control 2', [[2022, 10, 4], [2023, 1, 4], [2022, 12, 21], [[[2022, 11, 14], 0.06]], 0.025, 4], 0.881793], ['normal control 3', [[2039, 6, 4], [2039, 12, 4], [2039, 8, 31], [[[2039, 6, 27], 0.04], [[2039, 11, 1], 0.035], [[2039, 8, 29], 0.035], [[2039, 7, 21], 0.05]], 0.02, 2], 0.939891], ['normal control 4', [[2013, 10, 30], [2014, 4, 30], [2014, 1, 30], [[[2014, 3, 23], 0.06], [[2014, 3, 27], 0.07]], 0.02, 2], 0.505495]], [['regression accrual end inclusivity 1', [[2030, 9, 23], [2031, 9, 23], [2030, 11, 20], [[[2030, 12, 5], 0.04], [[2030, 11, 20], 0.08]], 0.025, 1], 0.39726], ['regression accrual end inclusivity 2', [[2017, 3, 21], [2017, 6, 21], [2017, 3, 31], [[[2017, 4, 6], 0.05], [[2017, 6, 17], 0.05]], 0.02, 4], 0.054348], ['partial repair probe 1', [[2020, 5, 30], [2020, 11, 30], [2020, 6, 18], [[[2020, 9, 12], 0.04], [[2020, 5, 19], 0.035]], 0.025, 2], 0.180707], ['partial repair probe 2', [[2040, 8, 27], [2041, 2, 27], [2040, 10, 20], [[[2040, 11, 9], 0.035], [[2040, 7, 31], 0.05], [[2040, 9, 25], 0.05]], 0.02, 2], 0.733696], ['normal control 1', [[2033, 9, 30], [2033, 12, 30], [2033, 10, 8], [], 0.025, 4], 0.054945], ['normal control 2', [[2028, 5, 31], [2029, 5, 31], [2029, 4, 16], [[[2028, 8, 18], 0.06]], 0.02, 1], 4.394521], ['normal control 3', [[2031, 6, 30], [2032, 6, 30], [2032, 6, 30], [[[2031, 10, 19], 0.04], [[2032, 2, 17], 0.035], [[2031, 6, 12], 0.06], [[2032, 1, 13], 0.07]], 0.02, 1], 4.710383], ['normal control 4', [[2020, 3, 1], [2020, 6, 1], [2020, 4, 29], [[[2020, 2, 12], 0.05], [[2020, 4, 6], 0.06], [[2020, 2, 6], 0.06], [[2020, 2, 28], 0.04]], 0.02, 4], 0.766304]], [['regression accrual end inclusivity 1', [[2023, 5, 7], [2024, 5, 7], [2023, 12, 4], [[[2023, 8, 29], 0.05], [[2023, 9, 12], 0.07], [[2023, 6, 15], 0.04], [[2023, 7, 9], 0.035]], 0.025, 1], 2.795082], ['regression accrual end inclusivity 2', [[2025, 8, 31], [2026, 8, 31], [2026, 5, 4], [[[2025, 10, 31], 0.06], [[2025, 12, 24], 0.06], [[2026, 5, 29], 0.035], [[2026, 3, 23], 0.06]], 0.02, 1], 3.375342], ['partial repair probe 1', [[2021, 3, 27], [2022, 3, 27], [2021, 5, 24], [[[2021, 12, 2], 0.07]], 0.025, 1], 0.39726], ['partial repair probe 2', [[2028, 9, 30], [2029, 9, 30], [2029, 8, 29], [[[2029, 8, 9], 0.03], [[2028, 11, 8], 0.05], [[2028, 11, 20], 0.04], [[2029, 8, 29], 0.08]], 0.025, 1], 3.467123], ['normal control 1', [[2022, 2, 28], [2023, 2, 28], [2022, 6, 8], [[[2022, 11, 1], 0.06]], 0.025, 1], 0.684932], ['normal control 2', [[2040, 5, 31], [2040, 11, 30], [2040, 6, 26], [[[2040, 8, 23], 0.03], [[2040, 7, 18], 0.03]], 0.025, 2], 0.177596], ['normal control 3', [[2013, 3, 24], [2014, 3, 24], [2014, 2, 4], [[[2013, 9, 17], 0.06], [[2013, 6, 18], 0.035]], 0.025, 1], 3.763014], ['normal control 4', [[2039, 6, 15], [2040, 6, 15], [2040, 1, 8], [[[2039, 9, 3], 0.03], [[2039, 10, 30], 0.04], [[2040, 5, 12], 0.07]], 0.02, 1], 1.669399]], [['regression accrual end inclusivity 1', [[2036, 9, 12], [2037, 3, 12], [2036, 12, 23], [[[2036, 11, 29], 0.03], [[2036, 10, 31], 0.04]], 0.02, 2], 0.790055], ['regression accrual end inclusivity 2', [[2021, 11, 19], [2022, 5, 19], [2022, 4, 30], [[[2021, 12, 23], 0.04], [[2022, 1, 3], 0.035], [[2022, 1, 21], 0.04], [[2022, 2, 8], 0.07], [[2022, 4, 30], 0.08]], 0.025, 2], 2.29558], ['partial repair probe 1', [[2032, 11, 24], [2033, 11, 24], [2033, 2, 3], [], 0.025, 1], 0.486301], ['partial repair probe 2', [[2021, 11, 7], [2022, 2, 7], [2021, 12, 17], [[[2021, 11, 28], 0.05], [[2022, 2, 12], 0.05], [[2022, 1, 19], 0.03]], 0.025, 4], 0.400815], ['normal control 1', [[2022, 1, 27], [2023, 1, 27], [2022, 4, 18], [[[2022, 5, 16], 0.03], [[2022, 2, 7], 0.06], [[2022, 11, 24], 0.03], [[2022, 4, 20], 0.05]], 0.025, 1], 1.226027], ['normal control 2', [[2028, 9, 1], [2029, 9, 1], [2028, 10, 7], [], 0.02, 1], 0.19726], ['normal control 3', [[2037, 2, 4], [2037, 5, 4], [2037, 4, 22], [[[2037, 4, 8], 0.035]], 0.02, 4], 0.491573], ['normal control 4', [[2015, 10, 1], [2016, 4, 1], [2015, 10, 24], [[[2016, 3, 16], 0.06]], 0.025, 2], 0.157104]], [['regression accrual end inclusivity 1', [[2027, 7, 31], [2028, 7, 31], [2028, 2, 28], [], 0.025, 1], 1.448087], ['regression accrual end inclusivity 2', [[2019, 11, 18], [2020, 2, 18], [2020, 2, 2], [[[2019, 11, 12], 0.06], [[2020, 1, 23], 0.06], [[2020, 1, 23], 0.03]], 0.02, 4], 1.23913], ['partial repair probe 1', [[2038, 4, 20], [2038, 10, 20], [2038, 6, 3], [[[2038, 7, 30], 0.03], [[2038, 10, 2], 0.07], [[2038, 6, 3], 0.08]], 0.02, 2], 0.240437], ['partial repair probe 2', [[2026, 3, 2], [2026, 6, 2], [2026, 5, 3], [[[2026, 4, 24], 0.03], [[2026, 3, 28], 0.07]], 0.02, 4], 0.728261], ['normal control 1', [[2014, 6, 27], [2015, 6, 27], [2015, 1, 25], [], 0.02, 1], 1.161644], ['normal control 2', [[2023, 5, 11], [2023, 11, 11], [2023, 8, 30], [[[2023, 9, 9], 0.03], [[2023, 5, 9], 0.035], [[2023, 8, 30], 0.08]], 0.02, 2], 1.055707], ['normal control 3', [[2010, 1, 30], [2011, 1, 30], [2010, 8, 26], [[[2010, 9, 1], 0.05], [[2010, 2, 10], 0.035], [[2010, 8, 26], 0.08]], 0.025, 1], 1.964384], ['normal control 4', [[2036, 10, 30], [2037, 4, 30], [2037, 2, 1], [[[2036, 10, 9], 0.07]], 0.025, 2], 1.807692]]]
for label, args, expected in fixtures[N-1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
regression accrual end inclusivity 1	`1.092896`	`1.092896`	Passed
regression accrual end inclusivity 2	`0.384615`	`0.384615`	Passed
partial repair probe 1	`1.546196`	`1.546196`	Passed
partial repair probe 2	`0.8`	`0.8`	Passed
normal control 1	`1.654795`	`1.654795`	Passed
normal control 2	`0.881793`	`0.881793`	Passed
normal control 3	`0.939891`	`0.939891`	Passed
normal control 4	`0.505495`	`0.505495`	Passed

SHA-256 / 4b59fa537f158a85b90a227bc958ded9796f64a93ab05cffebcf318676fc0b5f

Verification & scope

A deterministic toy contract stated explicitly in the contract field; no claim of conformance to any published convention text. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-29T14:46:52.821452+00:00.

Case digest / c523fe35af655a1c4fea5530391ca7199f9625f6e53c62d7720ef1eb266ee223