kibot

Adjusted vs Unadjusted Historical Data

Kibot distributes equity and ETF history in two parallel forms: adjusted and unadjusted. Adjusted data rewrites historical prices and volumes retroactively whenever a corporate action occurs, producing a continuous series suitable for technical analysis. Unadjusted data preserves the literal prices and share counts that printed on the tape on each trading day. Most backtests want adjusted data; most reconciliation and audit work wants unadjusted. This article lays out when each is appropriate, how Kibot's adjustments work, and why the same security can look different on Yahoo Finance.

Why use adjusted data

A few dozen stocks and ETFs are adjusted for splits and dividends every working day. When historical data is adjusted, every value before the event is rewritten by multiplying through a calculated ratio so that the series looks continuous across the event (per the Faq).

Adjusted data is what technical indicators and buy/sell signals should run against. If data is not adjusted for splits and dividends, indicators and any other calculated series will produce large price gaps on the dates immediately before and after the corporate action. Those synthetic gaps will trigger false buy or sell signals in any system that reacts to price changes.

If you need to compute percent returns relative to the prior day's close, unadjusted data will also give you incorrect results on dividend and split days. The return is only statistically meaningful when the series has been adjusted. A practical refinement is to compare the current day's data against the prior day's average price rather than close-to-close, and to prefer intraday bars over daily bars since the daily open and close carry their own ambiguities around the official open and close prints (see the discussion in Data format reference).

When to use unadjusted data

Unadjusted data is the right choice whenever you need to know the exact real-life price and volume on a specific date, for reconciliation with broker statements, reproducing historical news, auditing a print, or comparing against another vendor's unadjusted feed. All the values remain unchanged after corporate events like stock splits or dividends. Unadjusted history almost never changes, though it may occasionally be corrected if an error in the original feed is discovered (per the FAQ).

Kibot's adjustment method

Because most Kibot customers are already familiar with Yahoo Finance and frequently compare the two, Kibot adopts the Yahoo Finance adjustment method documented at help.yahoo.com/l/us/yahoo/finance/quotes/quote-12.html. Critically, Kibot adjusts for both splits and dividends, while Yahoo's main Close column adjusts for splits only, dividends are folded into a separate Adj Close column. That is the single most important fact to understand before comparing the two data sets.

Kibot's adjusted daily data will in most cases match Yahoo historical prices, with one exception: the daily closing price. In Kibot's files the close is the last transaction during the regular trading session. On Yahoo that value may instead be a transaction after the close. To find out what the exact official close price was on a specific date, refer to the unadjusted Close column in Yahoo Finance data.

Volume adjustment

Volume is adjusted in both intraday and daily files. In unadjusted history, volume is the literal number of shares that traded. In both adjusted and unadjusted data, the product of price and volume always yields the exact dollar amount exchanged during the period, the adjustment preserves the notional.

Concretely, consider a stock that traded at $10 on 100 shares, then underwent a 2-for-1 split:

Price,Volume
$10,100 <--- 2:1 split
$5,200

Total dollar amount is $1,000 in both cases.

Adjusted vs unadjusted: the difference in practice

Take two consecutive trading days. On the second day, the stock has a 2-for-1 split and the price drops from $20 to $10. The unadjusted series reports exactly what happened on the tape:

Unadjusted data:
$20 <--- 2:1 split
$10

The adjusted version rewrites every price before the split by the split ratio, so the series becomes continuous:

Adjusted data:
$10 <--- 2:1 split
$10

The adjusted series no longer shows a 50% one-day "drop", the corporate action has been absorbed into the history itself.

Comparison with Yahoo Finance

The most common support question in this area is: "Your daily data does not match Yahoo Finance historical prices." The answer is almost always that someone is comparing Kibot adjusted data against Yahoo's unadjusted Close column. The correct mapping is that Kibot adjusted should match Yahoo's Adj Close, and Kibot unadjusted should match Yahoo's Close. Yahoo adjusts only for splits, while Kibot adjusts for both splits and dividends, so the reliable comparison is always unadjusted-to-unadjusted. A fuller discussion lives in Comparing Kibot data to other sources.

Key Takeaways

  • Adjusted data rewrites prices and volumes retroactively for splits and dividends; unadjusted preserves the tape.
  • Use adjusted for technical indicators and return calculations, unadjusted introduces false gaps on split/dividend days.
  • Use unadjusted whenever the exact real-life price on a given date matters (reconciliation, audit, news).
  • Kibot adjusts for both splits and dividends; Yahoo's Close adjusts for splits only. Dividend treatment is the main source of mismatch.
  • Kibot adjusted ≈ Yahoo Adj Close; Kibot unadjusted ≈ Yahoo Close. Reliable comparisons use unadjusted on both sides.
  • Volume is adjusted in both intraday and daily files; price × volume is preserved across events.
  • Kibot's daily close is the last regular-session print, not the official post-close print; this can diverge from Yahoo's Close by a few cents.