Comparing Stock Performance with Indexed or Log-Scaled Prices

Summary

The document addresses how to make a chart of stocks with very different share prices easier to compare. One suggested method resets each stock to a common starting value, such as an index of 100, by dividing its prices by its initial price and applying the same scale factor. This makes subsequent chart movements show relative performance over the chosen period rather than differences in nominal share prices.

A second suggestion is to chart log prices. The response explains that logarithmic scaling makes equal vertical changes correspond approximately to equal percentage changes, helping compare growth across price levels. These are related but distinct choices: indexing emphasizes change from a common baseline, while log scaling transforms the price axis. The document offers conceptual guidance but no chart, empirical comparison, or discussion of choices such as dividends, split adjustments, or the selected start date.

Key ideas

• Index each stock to a common starting value to compare relative performance over a period.
• Logarithmic price scaling makes equal vertical movements correspond to similar percentage changes.
• Indexed prices and log prices address different aspects of cross-stock chart comparison.
• The choice of start date affects the relative-performance picture.

Tags

# Scale prices in multiple stocks for comparison


# Scale prices in multiple stocks for comparison












I have a chart with prices of multiple stocks.

The problem is that some prices are below 100 while others are above 1000, so the chart is really ugly.

How do I compare historical in multiple stocks, so the chart is still readable. For instace, if I add Berkshire Hathaway, I won't be able to read anything off my chart because of the large gap between prices of different stocks.

I wonder if that's what the logarithmic scale is for?

## Answer by Mats Lind (score 3, accepted)

https://quant.stackexchange.com/a/29639

Index all the stocks to 100 in the start of the period!That is mulitiply all stock prices with 100 and divide all stock prices with their price in the beginning of the time series!

## Answer by Malick (score 5)

https://quant.stackexchange.com/a/29641

Yes I would recommend you to plot the log of prices instead of prices. It will re-scale the data while preserving the hierarchy of prices, and more importantly it allows to compare easily the growth among several stocks because a vertical move of 0.01 corresponds to a 1% change of the price at any point in the figure (not matter the price level and the direction of the move).

See also this blog post and this SO question .

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.