Adjusted vs Unadjusted Stock Prices
Which price view to use, what each preserves, and how mixing them creates false backtest signals.
An unadjusted stock price is the quoted historical price for that session. An adjusted price transforms earlier observations for specified corporate actions, commonly splits and sometimes dividends, to support comparable return series. Use raw prices to reconstruct tradable quotes and adjusted series for defined return calculations; keep the policy explicit and never mix views silently.
Splits and dividends affect history differently
A stock split changes share count and quoted price without the same economic loss implied by a raw price drop. Split adjustment removes that discontinuity for returns. Dividend adjustment may also transform prior prices to represent total-return effects.
Providers can use different methods and update at different times. Record which actions and formula produced the adjusted series.
Choose by research task
Execution reconstruction needs prices a trader could have observed, with share quantities and corporate actions handled explicitly. Return and factor research often uses adjusted data, but event studies may retain both raw observations and an adjusted benchmark.
Volume and shares may require their own adjustment logic. Do not assume a price adjustment automatically makes every field economically consistent.
- Name the adjustment policy.
- Retain raw values.
- Version corporate actions.
- Recalculate derived returns after corrections.
Detect mixing errors
Large artificial returns around split dates, impossible fills, and mismatched chart values often indicate mixed series. Test known action dates and inspect both price and quantity transformations.
DataCedar carries the adjustment view in the dataset contract so a response cannot silently combine policies across symbols or intervals.
How DataCedar preserves the evidence
DataCedar separates acquisition from serving. Permitted source responses are retained with retrieval time and identifiers, normalized into DataCedar-owned tables, checked against expected coverage, and exposed through a stable versioned API. A collector can be replaced without changing the customer contract or making an upstream provider a runtime dependency.
Every research stream carries effective and known-at time where the distinction matters. Rights-restricted, unavailable, partial, stale, and genuinely empty states remain visible, so a backtest can fail closed and a buyer can see the product boundary before committing engineering time.
Key takeaways
- 01Raw prices reconstruct quoted history.
- 02Adjusted prices depend on a named policy.
- 03Keep corporate actions and raw values.
- 04Never combine raw entries with adjusted exits.
Query public company evidence through one stable API. Free Explorer tier, no card.
Start freeView pricing