Skip to content
All library documents

Adjusting Historical Stock Prices for Dividends and Corporate Actions

Article Quant Q&A · Author: user82582
Read the original

Summary

The document considers whether historical stock prices used in empirical modelling should be adjusted for dividends. It explains that a cash dividend, in an otherwise unchanged market, would theoretically reduce the stock price by the dividend amount when it goes ex-dividend. That reasoning assumes no other relevant events affect prices and applies specifically to cash dividends.

For historical analysis, adjusting only for cash dividends is incomplete because stock splits and other corporate actions also change quoted prices. The response therefore recommends using adjusted prices for general empirical work, such as the adjusted historical series supplied by the cited data source. The appropriate treatment still depends on the modelling objective. The discussion gives a conceptual recommendation rather than a detailed adjustment formula, and it cautions that stock dividends require different treatment from cash dividends.

Key ideas

  • A cash dividend can theoretically lower the stock price by its amount, assuming no other market events intervene.
  • Adjusting prices only for cash dividends overlooks splits and other corporate actions.
  • Adjusted historical prices are generally more suitable for empirical analysis that needs corporate actions accounted for.
  • The treatment depends on the research objective, and stock dividends differ from cash dividends.

Tags

EquitiesStatisticsBacktesting
Full text
# Adjusting open, highs and lows for past monthly stock prices?


# Adjusting open, highs and lows for past monthly stock prices?












I'm looking into modelling monthly stock prices and want to start off by using data from Yahoo Finance.

I know that the closing prices given there are adjusted for stock splits and dividends, but I'd like to use the given raw data and adjust it myself.

For modelling purposes, should I adjust the past prices for dividends? I've researched this and read mixed opinions expressed.

My thought process was that suppose XYZ closes at 100 before a dividend of 0.05 is implemented. On the next day, the stock should ideally open at 99.95 and automatically account for the dividend given out.

Is this right or am I really wrong? Sorry if this is a noob question.

## Answer by Quantopik (score 1)

https://quant.stackexchange.com/a/11221

The line of thinking is theoretically correct and it is right if you assume that:

- no other event happened during the trading day or in recent periods (if, for instance, one has a stock split recently, you will take into account also that and so on for all corporate events);

- The dividend is a cash-dividend (in the case you will have a stock-dividend things are slightly different);

As regards the first question, since you cannot assume those hypothesis in the real world, you have to consider adjusted prices, that, however, are provided by yahoo_finance (the last column in the historical data table). It depends on the kind of survey you have to do, but, generally, it is better to conduct an empirical analysis by using adjusted-prices because of the reasons above.

For the second one, you are in wrong because in that way you're adjusting prices only for cash-dividends, overlooking the other corporate finance events.

Also Here you can find an answer given on this site that you could find pretty useful.

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.