Home / Performance Metrics / Annualized Return
Return metricBeginner

Annualized Return

Annualized return is a period return scaled to a full-year rate by geometric compounding, computed for a per-period return r over p periods per year as (1 + r) raised to the power p minus one.

Quick answer: Annualized return is a period return scaled to a full-year rate by geometric compounding, computed for a per-period return r over p periods per year as (1 + r) raised to the power p minus one.

In simple words

Annualized return takes a return earned over a short slice of time, such as a month or a quarter, and asks what yearly rate it would represent if that pace continued and compounded. It lets you put a monthly result and a quarterly result on the same yearly footing. The catch is that projecting a short period out to a year assumes the pace repeats, which for a lucky month is often wildly optimistic.

Purpose

Annualized return exists to make returns measured over different period lengths comparable on a common yearly scale, and it underlies the annualisation of nearly every other metric, from Sharpe to volatility.

Professional explanation

Compounding a period return to a year

If a strategy returns r in one period and there are p such periods in a year, compounding that pace forward gives a yearly multiple of (1 + r) to the power p, so the annualized return is that multiple minus one. For a monthly return, p is 12; for a quarterly return, p is 4; for a daily return on a trading-day clock, p is about 252. The exponent is what distinguishes geometric annualisation from naive multiplication.

Why not simply multiply

Multiplying a monthly return by 12 ignores compounding and overstates the true annual figure whenever returns are positive, because it forgets that each month grows on the previous month's gains. A 2 percent monthly return is not 24 percent a year but (1.02) to the power 12 minus one, about 26.8 percent. Conversely for volatile or negative sequences, naive scaling misleads in the other direction. Geometric annualisation is the honest method.

Annualising a whole backtest versus a single period

There are two distinct calculations that are easy to confuse. Annualising a single period return with a known p, as above, projects that one period's pace forward. Annualising a multi-year backtest instead means finding the CAGR from its endpoints, which uses the (End ÷ Start) to the power (1 ÷ Years) form. When people say “annualized return” for a full multi-year track record, they almost always mean CAGR; the per-period compounding form is for scaling a single sub-period.

The extrapolation hazard

Annualising a short, favourable period is statistically dangerous because it assumes a small, noisy sample represents the long-run pace. A single strong quarter annualised can imply an absurd yearly rate that the strategy has no chance of sustaining. The shorter the base period, the larger the extrapolation error, and the more a lucky streak gets magnified. Annualised figures from under a year of data should be treated as illustrative arithmetic, not forecasts.

Consistency of the periods-per-year convention

The value of p must match the data's frequency and be used consistently everywhere in a research pipeline. Daily strategies commonly use 252 trading days; some use 365 calendar days for return series that accrue on weekends, such as certain funding or interest calculations. Mixing conventions, for example annualising returns with 252 but volatility with 365, produces internally inconsistent and non-comparable statistics across a strategy library.

Formula

Annualized return = (1 + r)^p − 1

r = the return earned in one period (as a decimal), p = the number of such periods in a year (12 for monthly, 4 for quarterly, about 252 for trading-day-daily). For a whole multi-year backtest, annualized return is instead the CAGR: (End ÷ Start)^(1 ÷ Years) − 1. Always keep p consistent with the data frequency across every metric.

Annualized return vs CAGR

AspectAnnualized return (period form)CAGR
InputOne period return r and pStart and end equity over Y years
Projects a pace forwardYes, assumes r repeatsNo, measures realised growth
Best forScaling a monthly or quarterly figureSummarising a full multi-year test
Extrapolation riskHigh for short base periodsLow, it is realised
Both are geometricYesYes

Practical example

Illustrative example (Indian market)

A Nifty momentum strategy returns 2.1 percent in a single month during the backtest. To annualise that one month, use r = 0.021 and p = 12: annualized return = (1.021)^12 − 1 ≈ 1.283 − 1 = 0.283, or about 28.3 percent. Note this is not a forecast; it merely states the yearly rate that one month's pace implies. Naively multiplying 2.1 by 12 would give 25.2 percent, understating the compounded figure, while treating 28.3 percent as a sustainable annual return would be reckless extrapolation from a single month.

Fund and PMS disclosures in India annualise returns for periods of one year or more but are required to show returns for periods under a year on an absolute (non-annualised) basis, precisely because annualising a few months of data would exaggerate the rate; a disciplined backtest should follow the same convention.

Advantages

  • Puts returns from different period lengths on one yearly scale
  • Uses geometric compounding, so it respects reinvestment
  • Underlies the annualisation of volatility, Sharpe and other metrics
  • Standard, well understood across the industry

Limitations

  • Annualising a short period assumes its pace repeats, which is often false
  • Extrapolation error grows as the base period shrinks
  • Blind to the path and to risk, like all point return metrics
  • Sensitive to the periods-per-year convention, which must be consistent
  • Easily abused to make a lucky month look like a spectacular yearly rate

Why it matters in practice

  • It is the bridge between raw period returns and comparable annual statistics
  • Its misuse on short samples is a common source of inflated marketing figures

Common mistakes

  • Annualising a single strong month or quarter and presenting it as a yearly expectation
  • Multiplying a period return by the number of periods instead of compounding
  • Mixing 252-day and 365-day conventions across different metrics
  • Confusing period-form annualisation with the CAGR of a full backtest
  • Annualising sub-year performance in disclosures where absolute figures are required
  • Ignoring that annualised figures still hide volatility and drawdown

Professional usage

Careful quants annualise only when the base period is long enough to be representative, and they clearly distinguish projecting a single period forward from computing the realised CAGR of a full track record. They fix the periods-per-year convention once, apply it uniformly to returns and volatility, and they refuse to annualise a few weeks of data into a headline. When they do annualise short periods for internal work, they label the figure as illustrative arithmetic, not a forecast.

Key takeaways

  • Annualized return scales a period return to a yearly rate by geometric compounding
  • For a period return r with p periods per year it is (1 + r)^p − 1
  • For a full multi-year backtest, annualized return means the CAGR
  • Annualising a short favourable period wildly overstates the sustainable rate
  • Keep the periods-per-year convention consistent across every metric

Frequently asked questions

What is annualized return?
Annualized return is a period return scaled to a full-year rate by geometric compounding. For a per-period return r over p periods per year, it is (1 plus r) raised to the power p, minus one.
How do I annualize a monthly return?
Take one plus the monthly return, raise it to the power 12, and subtract one. A 2 percent monthly return annualizes to about 26.8 percent, not 24 percent, because compounding is included.
Why not just multiply the period return by the number of periods?
Because multiplying ignores compounding and overstates positive returns, forgetting that each period grows on the previous one's gains. Geometric compounding, using the exponent, is the correct method.
Is annualized return the same as CAGR?
For a whole multi-year backtest, the annualized return is the CAGR. The per-period compounding form, (1 plus r) to the power p, is used to scale a single sub-period such as a month or quarter.
What value of p should I use for daily returns?
Usually about 252, the number of trading days in a year, if returns accrue only on trading days. Some series that accrue on all days use 365, but you must apply the choice consistently everywhere.
Why is annualizing a short period dangerous?
Because it assumes a small, noisy sample represents the long-run pace. A single strong quarter annualized can imply an unsustainable yearly rate, and the shorter the base period, the larger the extrapolation error.
Can I annualize a return of less than a year for a client report?
In Indian fund and PMS disclosures you must show sub-year performance on an absolute basis, not annualized, precisely to avoid exaggeration. A disciplined backtest should follow the same convention.
Does annualized return account for risk?
No. Like other point return metrics it is blind to volatility, sequence risk and drawdown. It must be paired with a risk or risk-adjusted measure to be meaningful.
How does volatility affect annualized return?
When you annualize a realized multi-period track record via CAGR, volatility drag lowers it below the arithmetic mean. When you project a single period forward, volatility is simply ignored, which is part of why the projection is unreliable.
Can annualized return be negative?
Yes. If the underlying period return is negative, compounding it forward produces a negative annualized rate, representing a yearly pace of loss.
What is the difference between annualized and cumulative return?
Cumulative return is the total gain over the whole period regardless of time, while annualized return expresses that on a per-year basis. Annualizing makes different horizons comparable; cumulative does not.
Should I annualize using arithmetic or geometric mean?
Geometric, because it respects compounding and reinvestment. The arithmetic mean overstates growth and is only used in specific contexts such as certain expected-return inputs, not for reporting realized annual performance.
Why do my annualized numbers differ between two tools?
Almost always because of a different periods-per-year convention, such as 252 versus 365, or arithmetic versus geometric annualisation. Aligning the conventions usually reconciles them.
Is a high annualized return from three months trustworthy?
No. Three months is far too short to represent a stable pace, and annualising it magnifies any luck. Treat such a figure as arithmetic illustration, never a forecast of what the year will deliver.

Voice search & related questions

Natural-language questions people ask about Annualized Return.

What does annualized return mean?
It is the yearly rate that a shorter-period return would represent if that pace kept up and compounded through the year.
How do I annualize a monthly return?
Add one to the monthly return, raise it to the twelfth power, then subtract one to get the yearly equivalent.
Is annualized return the same as CAGR?
For a full multi-year backtest, yes, they are the same idea; the compounding formula is mainly for scaling a single month or quarter.
Why is annualizing a few months risky?
Because a short lucky stretch gets magnified into an unrealistic yearly rate, so it should be treated as illustration, not a forecast.
Should I multiply or compound to annualize?
Compound it, because simply multiplying ignores that each period earns on the previous period's gains and overstates the result.
Can annualized return be negative?
Yes, if the underlying period return is negative, the annualized figure will be negative too, showing a yearly rate of loss.

Sources & references

    Last reviewed 11 July 2026. Educational content only — not investment advice. Markets and rules change; verify current conventions with SEBI, NSE/BSE and your broker.

    ← Previous
    Absolute Return
    Educational content only — not investment advice. Examples use illustrative numbers and simplified models. Backtested results are hypothetical and trading derivatives involves substantial risk. See our Risk Disclosure and SEBI Disclaimer.