Bug: Points System crashes with ValueError due to p.retire being a list (plus subsequent TypingError)

[lnguyen-kit]

Hi,

I am attempting to run the model using the "Points System" for pensions. I encountered a blocking error immediately when running the baseline.

1. The Primary Issue (ValueError) When p.retire is a list (representing the time path, length 400), the function PS_amount in pensions.py crashes at np.zeros(p.retire). It interprets the list as dimensions rather than the length of the array.

Traceback:
File "C:...\ogcore\pensions.py", line 412, in PS_amount
L_inc_avg_s = np.zeros(p.retire)
ValueError: maximum supported dimension for an ndarray is currently 64, found 400

2. Secondary Issue (TypingError) If I manually patch the code above to use int(p.retire[0]), I immediately hit a second error regarding g_y (growth rate). If g_y is a scalar (float), the Numba-compiled loop fails because it tries to access g_y[-1].

Traceback (after patching the ValueError):
numba.core.errors.TypingError: Failed in nopython mode pipeline (step: nopython frontend)
No implementation of function Function() found for signature:

getitem(float64, int64)
...
File "...\ogcore\pensions.py", line 888:
for s in range(S_ret):
L_inc_avg_s[s] = w[s] / np.exp(g_y[-1] * (u - s)) * e[s] * n[s]
^

Environment:

• OS: Windows 11
• Python 3.11 / Anaconda

Thanks for your help!

[jdebacker]

@lnguyen-kit Thanks for flagging this.

Related to Issue #1014