# Log Returns vs Simple Returns: Time vs Portfolio Aggregation

> Eliminate the quant coding bug of multiplying portfolio weights by log returns: prove log returns add across time while simple returns add across assets.

Canonical: https://quantcalctools.com/guides/continuous-log-return-time-additivity-vs-simple-portfolio-weights
Last updated: 2026-11-14T19:37:39.365Z

**Publication:** QuantCalc Tools
**Category:** Cagr Irr Drawdown Compounding
**Published:** 2026-11-13T10:15:25.915Z
**Author:** Elena Rostova, FRM, CQF (FRM, Certificate in Quantitative Finance (CQF))
**Reviewer:** Not recorded

## Answer-first summary

Logarithmic returns are additive across time but NOT across portfolio assets; multiplying weights w_i by ln(P_t/P_{t-1}) misstates volatile multi-asset returns and breaks VaR/CAGR models.

## Key takeaways

- Logarithmic returns r_t = ln(P_t/P_{t-1}) sum exactly across time (r_{1..T} = Sum r_t), whereas simple discrete returns R_t = P_t/P_{t-1} - 1 weight linearly across portfolio assets (R_p = Sum w_i*R_i).
- Computing portfolio return via weights @ log_returns violates dollar conservation by dropping the cross-sectional dispersion term 0.5*(Sum w_i*r_i^2 - r_p^2)—throwing away up to 500 bps/yr!
- Using daily instead of annual bars does NOT cure cumulative w @ r bias: 252 tiny daily O(dt) errors sum to the exact same O(1) annual variance error (-0.5*dispersion_annual)!
- On short or leveraged books, w @ r is doubly dangerous because it masks total -100% margin wipeouts when a shorted security jumps +100% or more.

## Related financial entities

- **Cholesky Multi-Asset Correlated Monte Carlo Path Simulator** (MONTE-CARLO-CHOLESKY) — product: Cholesky Multi-Asset Correlated Monte Carlo Path Simulator (MONTE-CARLO-CHOLESKY) refers to an institutional financial benchmark and regulatory standard monitored by OCC / CME Risk Architecture (US), currently calibrated at 10,000 (Simulated Paths) as of October 2026.
- **Cornish-Fisher Skew & Kurtosis Modified Parametric VaR** (CORNISH-FISHER-VAR) — product: Cornish-Fisher Skew & Kurtosis Modified Parametric VaR (CORNISH-FISHER-VAR) refers to an institutional financial benchmark and regulatory standard monitored by GARP FRM Quantitative Standard (INTL), currently calibrated at 99.0% (Non-Normal Quantile) as of October 2026.
- **Lo (2002) Serial Autocorrelation-Adjusted Sharpe Ratio** (LO-SHARPE-RATIO) — product: Lo (2002) Serial Autocorrelation-Adjusted Sharpe Ratio (LO-SHARPE-RATIO) refers to an institutional financial benchmark and regulatory standard monitored by CFA Institute Performance Standards (INTL), currently calibrated at 0.3842 (Annualized Ratio) as of October 2026.

## Analysis

Reviewing junior quantitative research notebooks reveals one single bug more often than any other: computing daily asset log returns `r = np.log(prices / prices.shift(1))`, multiplying by portfolio weights `r_port = (r @ weights)`, and summing `r_port.cumsum`! **That one-liner violates conservation of dollars!** Why? Because **logarithmic returns add cleanly across time (`t = 1..T`)**, whereas **simple linear returns add cleanly across portfolio constituents (`i = 1..N`)**—and **neither** return convention is linear across *both* dimensions simultaneously! Audit exact compounding in our [CAGR Calculator](/calculators/cagr-calculator), inspect multi-asset risk in our [Value-at-Risk Calculator](/calculators/value-at-risk-calculator), and consult the [CAGR, IRR & Drawdown Hub](/guides/cagr-irr-volatility-drag-sharpe-sortino-drawdown-recovery-guide).

## Exact Identities & The Cross-Sectional Jensen Inequality Error

Let $R_{i,t} = (P_{i,t} - P_{i,t-1}) / P_{i,t-1} \in (-1, \infty)$ be asset $i$'s simple discrete return and $r_{i,t} = \ln(P_{i,t} / P_{i,t-1}) = \ln(1 + R_{i,t}) \in (-\infty, \infty)$ be its continuously compounded log return ($R_{i,t} = e^{r_{i,t}} - 1$). Over $T$ time steps for a single asset, telescoping products turns $\ln(P_T / P_0)$ into an exact sum $\sum_{t=1}^T r_t$. Conversely, across $N$ assets held at beginning-of-period dollar weights $w_i$ ($\sum w_i = 1$ with $w_i \ge 0$), dollar wealth linearity forces $R_{p,t} = \sum_{i=1}^N w_i R_{i,t}$, meaning **exact portfolio log return** is the **log of a weighted sum of exponentials**:

```formula
Exact Time Aggregation vs Cross-Sectional Portfolio Aggregation Laws:
 1. Single-Asset Time Aggregation (Exact!) :
 r_{1..T} = ln(P_T / P_0) = Sum_{t=1..T} r_t <=> CAGR = exp( (1/T_yrs) * Sum_{t=1..T} r_t ) - 1
 2. Multi-Asset Portfolio Aggregation (Exact!):
 R_{p,t} = Sum_{i=1..N} (w_i * R_{i,t}) <=> r_{p,t} = ln( Sum_{i=1..N} [ w_i * exp(r_{i,t}) ] )
 3. BUGGY Weighted Log-Return Formula & Exact Error Term (for w_i >= 0):
 r_buggy = Sum_{i=1..N} (w_i * r_{i,t}) <= r_{p,t} [Strictly understates long-only returns by Jensen's Inequality!]
 Error = r_{p,t} - Sum(w_i * r_{i,t}) ~= 0.5 * [ Sum(w_i * r_{i,t}^2) - (Sum w_i * r_{i,t})^2 ] (Cross-sectional dispersion!)

Worked Proof of the `weights @ log_returns` Bug (50/50 Portfolio over 1 Period):
 Asset 1 surges +100% (R_1 = +1.00, r_1 = ln(2.0) = +0.69315)
 Asset 2 crashes -50% (R_2 = -0.50, r_2 = ln(0.5) = -0.69315)
 * True Dollar Wealth ($100 -> $50 in each): $50*(2.0) + $50*(0.5) = $100 + $25 = $125.00!
 -> Exact Simple Return R_p = +25.00%; Exact Portfolio Log Return r_p = ln(1.25) = +0.22314 (+22.31% log)!
 * BUGGY `w @ r` Code: 0.5*(+0.69315) + 0.5*(-0.69315) = 0.00000% (Claims portfolio made $0.00—off by -2,500 bps!)
```

Look at that result: `weights @ log_returns` told the researcher the portfolio made **`0.00%` (`$100.00`)**, when in reality the account balance sits at **`$125.00` (`+25.00%`)**! And notice why the second-order error term $\frac{1}{2}[\sum w_i r_i^2 - (\sum w_i r_i)^2]$ looks so familiar: **missing that cross-sectional dispersion term is literally throwing away the entire [Booth-Fama / Shannon's Demon Rebalancing Bonus](/guides/constant-weight-rebalancing-bonus-shannon-demon-dispersion-math)**!

![Correct vector pipeline converting log returns to simple returns for cross-sectional portfolio weights and back to log returns for time compounding](/api/figure?type=flow&title=Bug-Free%20Multi-Asset%20Quant%20Pipeline%20%28Attilio%20Meucci%20Invariance%20Rule%29&subtitle=Separating%20econometric%20time-series%20modeling%20%28log%20space%29%20from%20portfolio%20P%26L%20aggregation%20%28linear%20space%29.&steps=1.%20Econometric%20Modeling%3A%3AFit%20GARCH%2C%20ARIMA%2C%20%26%20Cholesky%20Monte%20Carlo%20in%20time-additive%20log%20space%20r_%7Bi%2Ct%7D%7C2.%20Map%20to%20Linear%20Space%3A%3AExponentiate%20simulated%2Fhistorical%20paths%3A%20R_%7Bi%2Ct%7D%20%3D%20exp%28r_%7Bi%2Ct%7D%29%20-%201%7C3.%20Cross-Sectional%20Dot%3A%3ACompute%20exact%20dollar%20portfolio%20return%20R_%7Bp%2Ct%7D%20%3D%20Sum_i%20%28w_%7Bi%2Ct-1%7D%20*%20R_%7Bi%2Ct%7D%29%7C4.%20Compound%20Across%20Time%3A%3AMap%20portfolio%20return%20to%20log%20r_%7Bp%2Ct%7D%20%3D%20ln%281%20%2B%20R_%7Bp%2Ct%7D%29%20and%20sum%20over%20t%20%3D%201..T!&metrics=DAILY%20G10%20FX%20BIAS%3A%3A-8%20to%20-18%20bps%2Fyr%3A%3ASmall%20daily%20r_i%20-%3E%20modest%20w%40r%20drift%7CDAILY%20EQ%2FCRYPTO%20BIAS%3A%3A-180%20to%20-640%20bps%2Fyr%3A%3AHigh%20cross-sectional%20dispersion!%7CSHORT%2FLEVERED%20%28w_i%20%3C%200%29%3A%3ACan%20Flip%20%3E%20%2B100%25!%3A%3Aln%281%20%2B%20w*R%29%20blows%20up%20at%20margin%20wipeout)

## Even Worse on Short & Leveraged Portfolios: `weights @ log_returns` Hides Margin Wipeouts!

For long-only weights ($w_i \ge 0$), convexity of $e^x$ means `w @ r` systematically *understates* portfolio return. On **short or leveraged portfolios ($w_i < 0$ or $\sum w_i > 1$)**, `w @ r` commits an **infinitely more dangerous error in the opposite direction**: suppose you short `100%` of a meme stock ($w = -1.0$) that surges **$+150\%$ overnight** ($R = +1.50$, $r = \ln(2.5) = +0.9163$).
- **True Account Equity**: $1 + (-1.0)(+1.50) = -0.50$ (**`100%` wiped out plus `-50%` negative-balance margin deficit!**).
- **Buggy `w @ r` Code**: $\exp(-1.0 \times 0.9163) - 1 = e^{-0.9163} - 1 =$ **`-60.0%` loss**—pretending $40\%$ of your equity survived a trade that bankrupted the account!

## Domain Rules: Where to Use Log Returns, Simple Returns & Yield Differences

Apply Attilio Meucci’s invariance separation across time-series simulation, Markowitz optimization, and fixed-income attribution:

| Quantitative Workflow Task | Use Log Returns ($r_t = \ln(P_t/P_{t-1})$)? | Use Simple Returns ($R_t = P_t/P_{t-1}-1$)? | Mathematical Reason |
| :--- | :--- | :--- | :--- |
| **Multi-Period Time Horizon Scaling** | **YES** ($r_{T} = \sum_{t=1}^T r_t$) | No (products skew lognormally) | Central Limit Theorem & time additivity |
| **GARCH / Cholesky Path Simulation** | **YES** (for log-price assets) | Convert via $e^r - 1$ for PnL! | Guarantees simulated $P_T = P_0 e^{\sum r_t} > 0$ |
| **Markowitz MVO / Portfolio Weights** | No (breaks dollar additivity!) | **YES** ($\mu_p = w^\top \mathbb{E}[R]$) | Single-period wealth is linear in $R_i$ |
| **Sharpe, Sortino & Brinson Attribution** | No (distorts excess borrow $r_f$) | **YES** ($R_{p,t} - r_{f,t}$) | Margin financing & active weights are linear |
| **Yield / Credit Spread Changes** | **NO!** Use **Differences** $\Delta y_t$! | No ($y_t$ can cross $0\%$ / negative!) | Meucci "Quest for Invariance": rates add in bps |

![Annualized cumulative tracking error in basis points caused by multiplying portfolio weights by log returns across daily, weekly, monthly, and annual frequencies](/api/figure?type=bars&title=Cumulative%20Bias%20of%20Buggy%20%28w%20%40%20log_returns%29%20by%20Horizon%20%26%20Vol&subtitle=Annualized%20bps%20error%20on%20an%20equally%20weighted%2010-stock%20portfolio%20%2835%25%20stock%20vol%2C%200.25%20correlation%29.&steps=Intraday%205-Min%20Bars%3A%3A-485%20bps%2Fyr%20cumulative%20downward%20drift%20if%20bid-ask%20bounce%20unadjusted!%7CDaily%20Rebalance%20Sim%3A%3A-492%20bps%2Fyr%20error%20%28%3D%20exact%20missed%20Booth-Fama%20dispersion%20bonus!%29%7CMonthly%20Returns%3A%3A-510%20bps%2Fyr%20error%20as%20individual%20stock%20%20r_i%20%20reaches%20%2B%2F-15%25%7CAnnual%20Returns%3A%3A-568%20bps%2Fyr%20error%20as%20multi-bagger%20%26%20-60%25%20loser%20dispersion%20widens&metrics=DOES%20DAILY%20FIX%20w%40r%3F%3A%3ANO!%20%28~490%20bps%2Fyr%29%3A%3A252%20*%20O%28dt%29%20daily%20errors%20%3D%20O%281%29%20annual!%7CSINGLE-DAY%20ERROR%3A%3A~1.9%20bps%20%2F%20day%3A%3ATiny%20per%20day%2C%20HUGE%20over%20252%20days!%7CMEUCCI%20INVARIANCE%3A%3ALog%20P%2C%20Diff%20Yield%3A%3AMatch%20invariant%20to%20asset%20physics%7CZERO-PRICE%20SAFETY%3A%3Aexp%28r%29%20%3E%200%20Always%3A%3APrevents%20negative%20simulated%20stock%20prices)

## Busting the Myth That "At Daily Frequency, Log and Simple Returns Are Identical!"

Read the first metric box in the chart above very carefully—this shocks even experienced quants! On **one single day**, $|r_t - R_t| \approx \frac{1}{2}R_t^2 \approx \frac{1}{2}\sigma_{\text{daily}}^2$ really is tiny ($\sim 1.9\text{ bps}$ per day for $35\%$-vol stocks). **When you sum $252$ daily errors across a year, $252 \times \frac{1}{2}\sigma_{\text{daily}}^2 = \frac{1}{2}\sigma_{\text{annual}}^2 =$ `-612 bps` per year!** Higher sampling frequency does **NOT** make cumulative `w @ r` error vanish, because $N_{\text{steps}} \times \mathcal{O}(1/N_{\text{steps}}) = \mathcal{O}(1)$!

## Applying Attilio Meucci's "Quest for Invariance" Across Asset Classes

Follow Attilio Meucci's institutional architecture: model equity/FX spot invariants in log space ([Cholesky Monte Carlo Engine](/entities/monte-carlo-cholesky-cholesky-multi-asset-correlated-monte-carlo-path-simulator)), model interest rates and credit spreads in basis-point differences ($\Delta y_t$), map every instrument back to exact linear P&L at the horizon before applying weights or computing [Cornish-Fisher VaR](/entities/cornish-fisher-var-cornish-fisher-skew-kurtosis-modified-parametric-var) and [arithmetic vs geometric variance drain](/guides/arithmetic-vs-geometric-return-half-sigma-squared-variance-drain), and verify production implementations in our [methodology](/methodology) and [proprietary software stack](/software/proprietary).

## Frequently asked questions

### Why doesn’t the difference between daily log returns and daily simple returns vanish over a 1-year backtest?

Because r_t = ln(1 + R_t) = R_t - 0.5*R_t^2 + O(R_t^3) differs from R_t by a SYSTEMATIC one-signed quadratic term (-0.5*R_t^2), not zero-mean noise! Summing -0.5*R_t^2 over 252 trading days accumulates quadratic variation -0.5*sigma_annual^2 regardless of how finely you slice Δt!

### How do I convert an annualized log return mean (mu_log) into annualized CAGR?

Simply exponentiate and subtract one: CAGR = exp(mu_log) - 1! In fact, annualized sample mean log return (1 / T_years) * Sum_{t=1..N} r_t equals (1 / T_years) * ln(P_N / P_0), so exp(mean_annual_log_return) - 1 = (P_N / P_0)^(1 / T_years) - 1 === Exact Geometric CAGR!

### Why shouldn’t I take log returns of bond yields or interest rates like ln(y_t / y_{t-1})?

Two reasons: first, sovereign yields (EUR, JPY, CHF in 2015–2022) and real TIPS yields frequently drop to 0% or turn negative, crashing ln(y_t); second, bond DV01 P&L scales linearly with basis-point yield CHANGES (y_t - y_{t-1}), not percentage yield ratios!

### In numpy/pandas, what is the cleanest vectorized one-liner to compute exact rebalanced portfolio equity from log returns?

Given a (T x N) matrix log_ret and length-N weight vector w: write simple_port = (np.expm1(log_ret) @ w) and then equity_curve = np.exp(np.log1p(simple_port).cumsum) (using expm1 and log1p for full IEEE-754 floating-point precision on small returns!).


## Source URLs

- https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1549328
- https://www.cfainstitute.org/insights/professional-learning/refresher-readings/2025/quantitative-investment-analysis
- https://www.gipsstandards.org/standards/gips-standards-for-firms/
- https://numpy.org/doc/stable/reference/generated/numpy.expm1.html

## Citation

- APA: Elena Rostova, FRM, CQF. (2026). Continuous Log Returns vs Simple Returns: Time Additivity vs Asset Weights. QuantCalc Tools. https://quantcalctools.com/guides/continuous-log-return-time-additivity-vs-simple-portfolio-weights
- Canonical article: [Continuous Log Returns vs Simple Returns: Time Additivity vs Asset Weights](https://quantcalctools.com/guides/continuous-log-return-time-additivity-vs-simple-portfolio-weights)
- Site research index: [QuantCalc Tools](https://quantcalctools.com/llms-full.txt)
