Abnormal Return Calculator
A stock rises 12% in a year and the headline writes itself. But if the market rose 9% over the same stretch, and the stock was riskier than the market, how much of that 12% was actually earned, and how much was simply the weather? Abnormal return is the number that answers that question: the part of a return that expectations cannot explain.
The Abnormal Return Calculator below subtracts the expected return from the actual return. You can enter the expected return directly, using a benchmark such as the market index return over the same period, or estimate it with CAPM from the risk-free rate, beta, and the market return. An optional event window field sums daily abnormal returns into a cumulative abnormal return (CAR), which is the standard output of an event study around news such as an earnings announcement.
Keep this concept separate from its close namesake, abnormal earnings. Abnormal earnings compare a company’s profit with the cost of its equity capital and feed into valuation. Abnormal return, this page, compares an investment’s actual return with a market-based expectation and feeds into performance measurement and event studies. Same surname, different subtraction, different data.
Quick Answer
Key takeaways
- Abnormal return = actual return minus expected return. The whole concept lives or dies on how honestly the expected return is estimated.
- Two expectation routes: a benchmark return entered directly, or CAPM, where expected return = risk-free rate + beta x (market return minus risk-free rate).
- When the expected return comes from CAPM, the single-period abnormal return is exactly the number finance calls Jensen's alpha.
- Event studies add daily abnormal returns across a window into the cumulative abnormal return (CAR). One day's abnormal return is noise; a CAR across many events, tested for significance, is evidence.
The formula, step by step
Measure the actual return. Use the same period and the same return definition (price only, or total return with dividends) that your expectation will use. Mixing the two quietly corrupts the subtraction.
Set the expectation. Either take a benchmark return directly, the market index over the same window is the common choice, or build the expectation with CAPM so the investment is only credited for beating what its risk level promised.
Subtract. Actual minus expected. Positive means the investment beat its expectation, negative means it lagged, zero means it delivered exactly what the model priced in.
For an event, add the days up. Sum the daily abnormal returns across the event window to get the CAR, the market's total verdict on the news.
Need the CAPM inputs themselves worked out first? The CAPM expected return calculator builds the expected return as its main job, and the beta calculator derives beta from price history. If you want the unadjusted picture for contrast, the absolute return calculator reports the raw return with no expectation subtracted, which is exactly the number this page starts from.
Worked examples, checked against this exact tool
Three runs of the calculator, each reproducible with the pre-filled defaults or the mode switch:
| Run | Inputs | Result |
|---|---|---|
| Benchmark (default) | Actual 12.00%, expected 9.00% | Abnormal return +3.00% |
| CAPM mode | Rf 2.00%, beta 1.2, market 10.00%, actual 15.00% | Expected 11.60%, abnormal return +3.40% |
| Event window | Daily abnormal returns 0.5, 1.2, -0.41 | CAR +1.29% over 3 days |
The CAPM run deserves a slow read, because it shows why the expectation route matters. The market risk premium is 10% minus 2% = 8.00%. A beta of 1.2 loads 1.2 x 8.00% = 9.60% of premium on top of the 2% risk-free rate, giving an expected return of 11.60%. The same stock judged against the plain 9% benchmark would have shown a +6.00% abnormal return on a 15% actual return. Judged against what its risk promised, it earned +3.40%. Both numbers come from this tool; the second is the honest one for a high-beta stock.
Event studies and cumulative abnormal return
Researchers use abnormal returns to measure how much an event was worth: an earnings announcement, a merger, a dividend change, a stock split. The recipe, set out in MacKinlay's classic survey of event studies in economics and finance, is to estimate the expected return from a quiet estimation window before the event, compute the abnormal return on each day around the announcement, and sum them over a short event window, commonly days -1, 0, and +1, because news leaks a little before and settles a little after.
That sum is the CAR, and the event window field on this calculator runs exactly that sum: 0.5 + 1.2 + (-0.41) = +1.29%. A CAR near zero says the market shrugged. A large CAR says the event carried real news. Two cautions separate a classroom exercise from a publishable study. First, the expectation model matters: the literature's workhorse is the market model, which fits each stock's own alpha and beta by regression, while CAPM imposes the theory's restrictions; the expected return models overview from the Event Study Tools research project walks through the trade-offs model by model. Second, a single CAR proves nothing on its own. Real studies average CARs across dozens or hundreds of events and test whether the average is statistically distinguishable from zero.
For the benchmark leg of the comparison, long-run index behaviour is the usual anchor: the S&P 500 returns calculator shows what the broad market actually delivered over the window you are studying, so the expected return you type in is grounded in the index's own history rather than a guess.
What competing calculators miss
We reviewed the top-ranking abnormal return and Jensen's alpha tools before rebuilding this page. Most do the core subtraction correctly. The gaps are in the framing around it.
Winning move 1: the event window is built in. The strongest competing tool computes single-period Jensen's alpha well, with a full CAPM decomposition, but it stops at one period. The academic audience for this keyword needs the multi-day CAR, the sum that event studies actually report. This page includes the event window field and prints the CAR next to the single-period result, +1.29% on the standard three-day example.
Winning move 2: both expectation routes on one screen. Thin tools force one model: either a benchmark box or a CAPM form. This calculator offers both behind one mode switch, so you can see how much of an apparent outperformance survives the move from a raw benchmark to a risk-adjusted expectation, +6.00% shrinking to +3.40% in the worked comparison above.
Winning move 3: the Jensen's alpha identity, stated plainly. Students meet Jensen's alpha in portfolio theory and abnormal return in event studies and rarely learn they are the same subtraction when CAPM supplies the expectation. This page says so in the takeaways and the FAQs, which saves the cross-referencing hunt through two sets of notes.
Winning move 4: an honest significance warning. Competing pages present a positive abnormal return as a verdict. This one states up front that a single positive reading is not proof of skill and that daily returns are noisy, which is the caution every textbook attaches to the method and most calculator pages quietly drop.
Frequently asked questions
Abnormal return is the difference between what an investment actually returned and what it was expected to return: actual return minus expected return. A stock that returns 12% when 9% was expected has an abnormal return of +3.00%. The expected return can come from a benchmark, such as the market index over the same period, or from a model such as CAPM.
Subtract the expected return from the actual return. Either enter the expected return directly from a benchmark, or estimate it with CAPM: expected return = risk-free rate + beta x (market return minus risk-free rate). Example: actual 12%, benchmark 9%, abnormal return = +3.00%. With CAPM inputs of a 2% risk-free rate, a beta of 1.2, and a 10% market return, the expected return is 11.60%.
It is the same number as the single-period abnormal return when the expected return comes from CAPM: actual return minus CAPM expected return. Portfolio theory calls it Jensen's alpha and uses it to judge managers; event studies call the daily version abnormal return and sum it into CARs. The arithmetic is identical.
CAR is the sum of daily abnormal returns over an event window. If a stock posts abnormal returns of 0.5%, 1.2%, and -0.41% on days -1, 0, and +1 around an announcement, the CAR is +1.29%. Event studies use it to measure the total price reaction to the news, with the market-wide movement stripped out.
Yes, and it is common. A negative abnormal return means the investment lagged what its benchmark or model said it should earn. A stock that returns 7% against a 9% expectation has an abnormal return of -2.00%, even though the raw return was positive. The sign compares performance with expectations, not with zero.
No. One period's abnormal return is mostly noise, because daily and monthly returns move for thousands of reasons the model does not capture. The number becomes meaningful only with repetition: many periods for a manager, or many events averaged together in an event study, plus a statistical significance test. Treat a single reading as a prompt for questions, not a conclusion.
Use a published beta from a data provider, or one estimated by regressing the stock's past returns on the market's returns. Beta scales the market risk premium: at a premium of 8%, a beta of 1.2 adds 9.60% of expected return, while a beta of 0.8 adds only 6.40%. A higher beta raises the bar the actual return has to clear, which is the point of the risk adjustment.
Add This Calculator to Your Website
References
- MacKinlay, A. Craig, "Event Studies in Economics and Finance," Journal of Economic Literature, 35(1), 1997, pp. 13-39 - the standard survey of event study method: estimation windows, expected return models, abnormal returns, and significance testing. MacKinlay (1997): Event Studies in Economics and Finance, JSTOR. Accessed October 2026.
- Abnormal return, reference overview - the definition of abnormal return as actual minus expected return and its use in event studies and performance evaluation. Abnormal Return: Definition and Uses. Accessed October 2026.
- Capital asset pricing model, reference overview - the CAPM relation E(R) = Rf + beta x (Rm minus Rf) used to estimate the expected return in the CAPM mode of this calculator. Capital Asset Pricing Model: Formula and Assumptions. Accessed October 2026.
- Jensen's alpha, reference overview - the definition of alpha as actual return minus the CAPM-predicted return, and why the measure is the same subtraction as the single-period abnormal return. Jensen's Alpha: Definition and Relation to Abnormal Return. Accessed October 2026.
- Fama-French Data Library, Dartmouth College Tuck School of Business (Kenneth R. French) - the standard source of factor and market return data used to estimate expected returns and betas in empirical work. Fama-French Data Library: Kenneth R. French, Dartmouth. Accessed October 2026.
- EventStudyTools, "Expected Return Models for Event Studies" - a research-project comparison of the mean-adjusted, market-adjusted, market model, and CAPM expectations used in event studies. EventStudyTools: Expected Return Models for Event Studies. Accessed October 2026.