Verification Suite

Loss disallowed, basis adjusted

Sell at a loss, repurchase within 30 days → the loss is disallowed and the replacement lot's cost basis increases by the disallowed amount.

Partial match (sell 100, rebuy 60)

Only 60 shares' worth of loss is disallowed; the remaining 40 shares' loss is deductible.

Wash sale tacks holding period

The original lot's holding period transfers to the replacement lot — a ST purchase can inherit LT status.

Disallowed losses skip netting

A wash-sale disallowed loss doesn't reduce your gains in the capital gains netting calculation.

Repurchase on day 31 is not a wash sale

Selling at a loss and repurchasing 31 days later falls outside the IRS 30-day window, so the loss remains deductible.

Buy before loss sale still triggers wash sale

Repurchasing within 30 days before a loss sale also triggers wash sale treatment. The disallowed loss is deferred into replacement basis.

Substitute ETF can still trigger wash sale

Selling SPY at a loss and buying a substantially identical substitute (VOO) within 30 days still disallows the loss when both map to the same identity group.

Non-identical replacement does not trigger wash sale

Selling SPY at a loss and buying QQQ in the window does not trigger wash-sale treatment when they map to different identity groups.

1099-B short-term gain (no wash sale)

Simple buy then sell in under one year with no replacement purchase. Broker-style 1099-B should report a short-term gain and zero wash adjustment.

1099-B long-term gain (no wash sale)

Holding period exceeds one year, so the realized gain must land in the long-term 1099 bucket.

1099-B full wash sale (loss fully disallowed)

Sell at a loss, then fully replace within 30 days. Broker-style output should show a wash adjustment that eliminates the currently deductible loss.

1099-B partial wash sale (loss partially disallowed)

Sell 100 shares at a loss, then repurchase only 40 shares. Only part of the loss is disallowed, leaving a deductible remainder.

1099-B cross-year wash sale (disallowed in sale year)

Loss sale in late December with replacement in early January still falls within the 30-day window. The sale-year 1099-B row should carry the wash adjustment.

1099-B mixed short-term and long-term rows in one year

A single sale can consume both long-held and recently acquired lots. Broker-style output should separate proceeds/basis/gain into LT and ST buckets for the same tax year.

1099-B aggregates multiple short-term sales

Multiple short-term sales in the same year should aggregate cleanly to broker-style annual short-term totals.

NIIT kicks in above $200k

$180k income + $100k LT gain → MAGI $280k. NIIT = 3.8% × min($100k investment income, $80k over threshold).

0% LT rate for low income

$0 ordinary income + $30k LT gain → $0 federal LTCG tax. The 0% bracket extends to ~$49k for Single filers.

Capital gains stack on ordinary income

The same $40k LTCG is taxed at 0% with $0 wages, but at 15% with $300k wages. Capital gains "stack" on top of ordinary income.

Qualified dividends can be taxed at 0%

$30k qualified dividends with no wages can remain in the 0% LTCG/qualified-dividend bracket for Single filers.

Ordinary dividends are taxed at ordinary rates

$30k ordinary (non-qualified) dividends with no wages still produce federal tax because they are taxed like short-term income.

CA state tax on $1M income > $100k

$1M taxable income in California (top bracket 13.3%). State tax should exceed $100k.

MFJ pays less tax than Single on $200k

$200k in short-term gains: Married Filing Jointly brackets are wider, so total federal tax is lower than Single.

Texas has no state income tax

Even at $1M taxable income, Texas state income tax remains $0 in the state-tax engine.

Taxable monthly rotation creates visible tax drag

A deterministic monthly rotation between two appreciating assets can look fine pre-tax while realizing enough short-term gains to materially reduce after-tax wealth in a taxable account.

Tax-free account removes the drag from the same trades

The exact same monthly rotation creates no tax drag when the account type is tax-free. This isolates account treatment from the trading pattern itself.

TLH harvests losses without breaking cooldown rules

When a taxable position falls enough to trigger tax-loss harvesting, ArthaPilot should realize the loss, rotate into the substitute ETF, and avoid immediately buying the original ticker back during the cooldown window.

Cross-year wash sale restates the prior-year tax bill

A loss realized near year-end and repurchased within 30 days in the next tax year should create wash-sale records and change the prior year's settled tax bill.

Ordinary dividends create drag in taxable accounts

The same bond-fund dividend stream should produce tax drag in a taxable account and none in a tax-free account, isolating dividend taxation from the market path.

ISO spread creates AMT liability

$90k ISO spread + $150k wages. The ISO spread is an AMT preference item (not regular income) → AMT exceeds regular tax.

NSO spread pushes over Medicare threshold

$195k wages + $10k NSO spread = $205k → exceeds $200k threshold → Additional Medicare = 0.9% × $5k = $45.

Small ISO spread can still owe zero AMT

A modest ISO spread does not always generate AMT. If AMTI stays near the exemption amount, AMT owed can remain zero.

Leveraged drawdown > 100%

Price path [100, 50, -10] (leveraged blowup). Max drawdown = 110%, not clamped to 100%.

Returns capture overnight gaps

2019 closes at 100, 2020 opens at 110, closes at 132. The 2020 return is 32% (from prior close, not open).

Multiple drawdowns — longest wins

Two drawdowns (1 day and 6+ days) with recovery between. The engine reports the longest one.

Sharpe ratio on near-constant returns

252 daily returns of +0.04% each (≈10% annualized) with tiny noise. rf=4%. The Sharpe ratio should be high and positive.

Sortino > Sharpe for positive-skew returns

Returns with positive skew (many small gains, rare small losses). Sortino exceeds Sharpe because downside volatility is less than total volatility.

Calmar ratio on known growth path

Value path from 100→120 over 252 trading days with a 10% drawdown. Calmar = CAGR / max drawdown.

CAGR of doubling in three years ≈ 26%

$100 → $200 over 756 trading days (3 years). CAGR = (200/100)^(1/3) - 1 ≈ 26%.

Withdrawal clamping on depleted portfolio

A portfolio crashes 50% in month 2, wiping out most of the withdrawal budget. ArthaPilot clamps each withdrawal to the available balance, so the total withdrawn reflects what the portfolio could actually pay — not a phantom fixed amount.

Average drawdown: full curve vs. underwater-days only

A 21-value curve: 19 all-time highs, then one trough at −10%, then recovery. One method averages only over the single drawdown observation (−10%). ArthaPilot averages the drawdown depth at every point, so the 19 ATH days (0% drawdown each) pull the average down to ~0.5%.

Fees reduce optimal leverage

Same μ/σ but with fees: optimal leverage (net) is lower than optimal leverage (gross).

Mixed 1x + 3x = 2x effective

$5k SPY (1x) + $5k UPRO (3x): total exposure $20k / total value $10k = 2.0x effective leverage.

Volatility drag on 2x LETF

Alternating +10%/−10% daily returns for 252 days. The 2x leveraged ETF loses more than the unleveraged version due to volatility drag.

Circular wrapping preserves history

Block bootstrap with 12-month history wraps around the end seamlessly. All simulated values come from the original history.

Depleted portfolio never goes negative

-30%/month returns + $60k/year withdrawal on $100k. Once the portfolio hits $0, it stays there (no borrowing).

Percentage withdrawal scales with portfolio

Moderate negative returns (-0.5%/month) with 10% annual percent-of-portfolio withdrawal. Portfolio shrinks but never hits $0.

Fixed withdrawals can deplete a flat portfolio

With 0% monthly returns and a fixed $12k annual withdrawal, a $100k portfolio is eventually depleted. This illustrates pure spending-rate risk.

Sequence risk with identical average returns

Two retirement paths use the same return set and identical average return, but in reverse order. Early losses can deplete capital even when long-run averages match.

Min-vol beats equal weight

Three assets with different volatilities. The min-volatility optimizer finds a portfolio with lower vol than naive equal-weight.

Max Sortino with asymmetric returns

One asset has positive skew (many small gains, rare losses), the other negative skew. Max Sortino should favor the positive-skew asset.

Leverage blocked for risk parity

risk_parity + allow_leverage=True is rejected because leverage breaks the equal-risk-contribution invariant.