Detecting Stock Splits in Historical Price Data
Summary
The document explains how to determine whether historical stock prices reflect splits. Its central point is that split-adjusted prices have already been rescaled, so the split itself cannot be identified by inspecting that adjusted series alone. Unadjusted prices may instead show a sharp drop around the split date, which can be checked against a company’s known split history.
The Apple example describes a seven-for-one split in 2014, when the share count rose and the per-share price fell by roughly the same ratio. The responses also suggest consulting historical split calendars or financial data providers to verify dates and compare them with the dataset. This is a practical identification approach rather than a complete data-validation procedure: price movements can have other causes, and the document does not explain how providers define adjusted close or handle dividends. Its guidance is therefore to cross-check candidate price discontinuities against corporate-action records.
Key ideas
- Adjusted prices do not reveal whether a stock split occurred because the series has already been rescaled.
- Unadjusted prices may show a substantial per-share price change around a split date.
- Compare suspected split dates and prices with independent corporate-action records.
- A split changes shares outstanding and per-share price without, by itself, changing the company’s total value.
Tags
Full text
# Identify stock split from historical price data # Identify stock split from historical price data I have price series data with open, high, low, close and adjusted close for different companies. I am not sure whether this data is adjusted for stock splits or not and would like to know how to check that. ## Answer by Bob Baerker (score 3) https://quant.stackexchange.com/a/38109 Only unadjusted data will permit you to see the stock split. You cannot look at adjusted data and tell if a split occurred. The are a variety of sites that list stock splits (Google: "Stock Split Calendar"). Fidelity provides a reliable one on a monthly basis. You'll have to capture it a month at a time of you want a historical list: https://eresearch.fidelity.com/eresearch/conferenceCalls.jhtml?tab=splits&begindate=2/1/2018 ## Answer by Stoockbroker (score 0) https://quant.stackexchange.com/a/38103 Stock splits makes the company shares more marketable and in this case doesn't affect their seed capital. It's rather book trick. So, first of all you should find a company that have made a split and compare it stock price day before and after that operation. This scenerio should be easy to see because stock splits are make by high-priced company so their market price (per share) after it is rather much lower than before. To be more reliable you can check it through some financial data provider as I mentioned below and compare it to your dataset. Here's some example. Apple (AAPL) on 9 June, 2014 made split 7:1, so total shares on market increased (7x times), but price per share was divided by 7. In this example price per share before this event occured was over \$640 and after not much over $90. If some company make operations like splits, dividends, etc you can check at google / yahoo finance. There also would be a date when that events occured.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.