সেন্সিটিভিটি অ্যানালাইসিস

def pw_of_annuity_factor(i, n):
    return ((1 + i) ** n - 1) / (i * (1 + i) ** n)

def project_pw(cost, revenue, expense, n, i):
    a = revenue - expense
    pa = pw_of_annuity_factor(i, n)
    return -cost + a * pa

# বেস কেস
COST = 100000
REVENUE = 30000
EXPENSE = 12000
N = 8
I = 0.10

base_pw = project_pw(COST, REVENUE, EXPENSE, N, I)
print(f"বেস কেস PW = ${base_pw:,.2f}")
print()

def sensitivity_sweep(var_name, cost, revenue, expense, n, i):
    print(f"ভেরিয়েবল: {var_name}")
    print(f"{'পরিবর্তন':>10} | {'নতুন মান':>14} | {'PW':>14}")
    print("-" * 46)
    pw_values = []
    for pct in [-0.30, -0.20, -0.10, 0.0, 0.10, 0.20, 0.30]:
        c, r, e = cost, revenue, expense
        if var_name == "cost":
            c = cost * (1 + pct)
            new_val = c
        elif var_name == "revenue":
            r = revenue * (1 + pct)
            new_val = r
        else:
            e = expense * (1 + pct)
            new_val = e
        pw = project_pw(c, r, e, n, i)
        pw_values.append(pw)
        print(f"{pct*100:>+9.0f}% | ${new_val:>12,.2f} | ${pw:>12,.2f}")
    swing = max(pw_values) - min(pw_values)
    print(f"PW সুইং (সর্বোচ্চ - সর্বনিম্ন) = ${swing:,.2f}")
    print()
    return swing

swing_revenue = sensitivity_sweep("revenue", COST, REVENUE, EXPENSE, N, I)
swing_expense = sensitivity_sweep("expense", COST, REVENUE, EXPENSE, N, I)
swing_cost = sensitivity_sweep("cost", COST, REVENUE, EXPENSE, N, I)

swings = {"রেভিনিউ": swing_revenue, "এক্সপেন্স": swing_expense, "ইনিশিয়াল কস্ট": swing_cost}
for name, s in sorted(swings.items(), key=lambda x: -x[1]):
    print(f"{name:>16}: PW সুইং = ${s:,.2f}")

most_sensitive = max(swings, key=swings.get)
print()
print(f"সবচেয়ে সংবেদনশীল ভেরিয়েবল: {most_sensitive} (সবচেয়ে বড় PW সুইং)")