Adjusted close vs close is a question about what a historical price series represents. A raw close records a closing price under the provider’s market-data definition. An adjusted close restates historical values for specified events, often splits and cash dividends, so comparisons can account for changes that a simple price ratio would otherwise mix together.
The word “adjusted” is not a complete methodology. Some series adjust only for share splits; others also incorporate distributions. The result can depend on the adjustment formula, reference date, event coverage and correction policy. Read the field description before treating two columns as interchangeable.
This guide uses a fictional three-for-one split and cash dividend to explain the arithmetic. It applies to researching stock data in India, Pakistan or elsewhere without assuming that every provider or exchange uses the same conventions.
Adjusted close vs close: identify the field first
A file can contain raw open, high, low and close alongside a separate adjusted-close column. The presence of that column does not mean the other fields were adjusted too. Alpha Vantage’s daily-adjusted documentation explicitly describes raw OHLCV fields, adjusted close and split/dividend events together in one response.
Start with the raw field’s meaning. An official closing price, a final trade, a settlement value and a last available quote can be different records. Then inspect what the adjusted field includes: splits, cash dividends, other distributions or some specified subset.
A historical adjusted value is expressed on the provider’s chosen comparison basis. It need not be a price at which a transaction occurred on that historical date. That does not make it erroneous; it means its purpose differs from reproducing the original quotation.
Keep the original and adjusted fields in separate columns. Replacing one with the other under an unchanged “close” heading makes later verification unnecessarily difficult. A reader should be able to tell which series fed a return calculation or chart.
For a concrete naming contrast, Sharadar distinguishes Close, which already reflects splits and stock dividends, from CloseAdj, which also accounts for cash dividends and spin-offs, and CloseUnadj, which preserves the unadjusted field. These are vendor-specific names, not universal definitions. If an export labels a column only “close,” retain the provider and original field identifier before renaming it for a worksheet. Otherwise a colleague may mistakenly adjust a series that has already been adjusted.
Follow one share through a three-for-one split
Begin with one hypothetical share valued at 240 rupees immediately before a three-for-one split. Assume that immediately afterward there are three shares valued at 80 each, with no other price movement. Total value remains 240 in this simplified example.
A raw close comparison of 80 ÷ 240 − 1 gives approximately −66.67%. That ratio ignores the fact that the original holding now contains three shares. It cannot describe the complete change in the holding’s value across the split.
To express the earlier price on the later share basis, divide 240 by three. The split-adjusted earlier close becomes 80, matching the later 80. The comparable price ratio is now 80 ÷ 80 − 1 = 0%.
The calculation changes the unit of comparison; it does not create value. Three-for-one means three new shares for each old share in this example. Always read the event’s ratio convention carefully, since a field might store new shares per old share, the inverse, or another adjustment representation.
Real market prices can move around a split for other reasons. The assumption that 240 becomes exactly three times 80 isolates the mechanical share-count change so the example can be checked. It is not a prediction about a company’s price after an announcement.
Add a cash dividend without losing the cash flow
Next suppose the three post-split shares each have a cash dividend entitlement of 2 rupees. For the simplified ex-dividend observation, assume the share price moves from 80 to 78 and nothing else changes. The shares are worth 234 and the distribution entitlement is 6, still totaling 240.
A split-only price series reports 78 ÷ 80 − 1 = −2.5% over that step. A return calculation that also recognises the 2-per-share distribution gives (78 + 2) ÷ 80 − 1 = 0%. These answer different questions.
One backward-adjustment method divides earlier prices by (ex-dividend close + dividend) ÷ ex-dividend close. Nasdaq Data Link’s Sharadar documentation describes this method for its adjusted historical series. In our example the divisor is 80/78, so the equivalent multiplier is 78/80 = 0.975.
Apply that multiplier to the split-adjusted earlier 80: 80 × 0.975 = 78. The dividend-adjusted history now shows 78 before and after the simplified distribution event. This method’s endpoint ratio incorporates the stated dividend treatment; another provider’s method must be checked on its own terms.
The ex-dividend event used in a return series is not necessarily the date cash arrives in an individual account. Record the event dates separately; our ex-dividend and record-date guide distinguishes entitlement from payment timing. Our arithmetic treats a distribution entitlement as part of the measurement and makes its assumed reinvestment explicit below.
Compare adjusted close vs close on a common basis
Finally, let the next raw close rise from 78 to 81 with no new corporate action. Using the final observation as the unchanged reference, the four-stage series is:
- Before split: raw close 240; split-adjusted close 80; split-and-dividend-adjusted close 78
- After split: raw close 80; split-adjusted close 80; split-and-dividend-adjusted close 78
- Ex-dividend observation: all three price fields 78
- Next close: all three price fields 81
The earliest fully adjusted value is 240 × one-third × 0.975 = 78. The two factors perform different jobs: one changes the share basis, while the other accounts for the specified distribution convention.
Compare the full-period ratios. Raw closes give 81 ÷ 240 − 1 = −66.25%, which ignores the share-count change and dividend. Split-adjusted prices give 81 ÷ 80 − 1 = 1.25%, representing the price change on a consistent share basis. The illustrated split-and-dividend-adjusted series gives 81 ÷ 78 − 1, approximately 3.8462%.
None of these figures should be copied into a report without its label. “The return was 3.8462%” leaves out assumptions that distinguish it from another valid cash-flow calculation.
Explain why the adjusted return can differ from cash held
Suppose the 6-rupee distribution is retained as cash. At the final price of 81, the three shares are worth 243. Add the 6 cash and final value is 249. Relative to the starting 240, the holding’s simple gain is 249 ÷ 240 − 1 = 3.75%, ignoring all costs and taxes.
Instead assume the 6 is reinvested at 78, with fractional shares allowed and no costs. It purchases 6/78, or about 0.076923 additional shares. The resulting 3.076923 shares are worth approximately 249.230769 at 81. Relative to 240, that is approximately 3.8462%, matching the illustrated adjusted-price ratio.
The difference comes from the additional shares participating in the later rise. It is an assumption about the constructed series, not evidence that every shareholder received that outcome. Actual reinvestment timing, eligibility, fractional-share handling and charges can differ.
CRSP’s calculations and index-methodology guide illustrates why this detail matters: its documented return calculations include distributions and specify reinvestment conventions. The broader lesson is to use a dataset’s actual return methodology rather than treating “adjusted” as a promise of personal after-cost performance.
For the broader distinction between endpoint price change and a distribution-inclusive series, see price return versus total return. Here the narrower task is to reconstruct the data fields and their adjustment factors before choosing a return label.
Understand why an old adjusted value can change
A later event can cause backward-adjusted historical levels to be restated. If a new split occurs, older prices may all be expressed on another share basis. A corrected distribution amount can also revise historical factors. Keep the data retrieval date and methodology version alongside an exported series.
A revised level is not automatically evidence that the original closing transaction was corrected. Compare the raw-close field and the event record separately. The provider may have changed the adjustment basis while leaving the underlying quote untouched.
There is also an important qualification for historical research. Multiplying every price in a past window by the same positive factor leaves percentage changes within that window unchanged. Yet absolute price thresholds, share quantities and comparisons with unadjusted series can change. Do not assume that every backward adjustment either creates or eliminates every timing problem.
For a study of information available on an earlier date, preserve contemporaneous inputs where needed and document subsequent events or revisions. The look-ahead bias guide explains that separate availability-time audit. Decide which fields the study actually uses before assessing the effect of a later restatement.
Use an event ledger to verify the series
Keep a short ledger containing security identifier, listing, currency, event type, event date, ratio or distribution amount, raw prices, adjustment factor, adjusted prices and retrieval date. Verify the issuer or exchange event notice when a large historical step looks unexplained; the guide to verifying market news provides a useful source-checking sequence.
Check field consistency before calculating a price range or indicator. Combining adjusted close with raw high and low can create relationships that did not exist in either complete series. Volume and share-count adjustments also need their own definitions; a price adjustment does not automatically establish how every other field was treated.
An unavailable event or close remains a missing input, not a zero-value event. Use missing market data versus zero to keep those cases separate. Record unresolved assumptions and later corrections in a market research journal.
The useful final question in an adjusted close vs close comparison is specific: does this field and its adjustment policy match the comparison I am making? For historical quote reconstruction, comparable price changes and distribution-aware returns, the required records may differ.
Adjusted close vs close: frequently asked questions
What is the basic difference between close and adjusted close?
A raw close preserves the provider’s closing-price observation, while adjusted close restates historical values for the events specified by that dataset. Inspect the definitions because “close” may already be split-adjusted in some products.
Does adjusted close always include dividends?
No universal label guarantees that. Some series are split-only; others incorporate cash dividends and additional events. Check the data dictionary, event coverage and methodology before interpreting an adjusted-price ratio as a distribution-aware return.
Was an adjusted historical price actually traded?
It may not have been. Backward adjustment expresses an old observation on a later comparison basis. The adjusted number can be useful for analysis while differing from the raw price recorded on that historical date.
Does a stock split automatically create a loss?
A share-count change must be considered with the per-share price. In the isolated example, one share at 240 becomes three at 80, preserving 240 of value. Actual market movements around a split remain separate from that mechanical relationship.
Why can adjusted close change months after the date shown?
Later corporate actions or corrected event data can change cumulative backward-adjustment factors. Save the retrieval date and compare raw prices separately. A restated adjusted level need not mean that the original traded close was wrong.
Can I use adjusted close with unadjusted high and low?
That mixes different bases unless the method explicitly accommodates them. An adjusted close may even fall outside the raw day’s range. For an indicator using several price fields, verify that all required inputs share a consistent adjustment convention.
Does adjusted return equal my account’s return?
Not necessarily. A constructed series may assume gross distributions, specific reinvestment timing and fractional shares. An account can have different transactions, cash balances, charges and other deductions. The article’s example deliberately separates cash retained from assumed reinvestment.
Is adjusted close the same as inflation-adjusted price?
No. Corporate-action adjustment and inflation adjustment address different changes. A series adjusted for splits and dividends is not automatically expressed in constant purchasing-power units, nor is it automatically converted to the reader’s home currency.
What if the company has a rights issue or spin-off?
Inspect the provider’s event-specific methodology and the issuer’s terms. The simple split and cash-dividend factors here do not establish the correct treatment of every corporate action. Do not invent a factor from an unexplained chart gap.
Which fields should I keep in a research export?
Retain raw close, adjusted close, event details, instrument identifier, currency, timestamps and retrieval date when available. Label any return formula and reinvestment assumption. Keeping the underlying records makes unusual steps and later revisions easier to investigate.
Explore CrowdSent and its current membership options for the wider service. This educational reconstruction does not claim that membership supplies adjusted-price exports, corporate-action files or account-return calculations.
Primary documentation checked 24 October 2026, including the July 2026 CRSP methodology guide.
Educational information only. All numerical examples are hypothetical. This article provides no investment recommendation or forecast.