Converting Kenneth French Library Risk-Free Rates from Percentages
Summary
The note clarifies the scale of risk-free rate values in the Kenneth French Data Library. Values such as 0.43 and 0.38 are expressed as percentages, rather than as decimal rates. To use them in calculations that expect decimal form, divide each value by 100; for example, 0.43 becomes 0.0043 and 0.38 becomes 0.0038.
This conversion matters when combining the series with returns or other rates expressed as decimals, since using the percentage figures directly would put the inputs on different scales. The response is limited to the unit conversion and does not discuss the dataset’s frequency, date conventions, or how to annualize or otherwise transform the rate. Verify that other inputs in a calculation use compatible units before comparing or combining them.
Key ideas
- The risk-free rate values in the cited library are percentage values.
- Convert a percentage value to decimal form by dividing it by 100.
- For instance, values of 0.43 and 0.38 become 0.0043 and 0.0038.
- Check that rates and returns use compatible units before combining them.
Tags
Full text
# Clarification on Risk-Free Rate Data from Kenneth French Data Library # Clarification on Risk-Free Rate Data from Kenneth French Data Library I noticed that the risk-free rate data from the Kenneth French Data Library seems unusual. Typically, the risk-free rate is a small number (e.g., around 0.3–0.4), but the dataset provides values such as 0.43 and 0.38, which appear relatively large. Could you please clarify whether these values should be used directly, or if I need to apply an additional conversion to the raw data before proceeding? Thank you very much. ## Answer by mehman (score 0) https://quant.stackexchange.com/a/84079 Those are percentage values, hence you should divide them by 100. E.g. 0.0043 or 0.0038.
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.