Why Fama–French Stock Return Data Begin in 1963

Summary

The note explains a possible reason Fama and French begin their stock return analysis in 1963: earlier COMPUSTAT records may be affected by substantial selection bias. The pre-1962 sample reportedly overrepresents large companies that had already been historically successful, so results based on those records could give a distorted picture of the broader stock universe.

The explanation is attributed to a discussion in another paper by the same authors. The note does not provide the underlying data analysis, quantify the bias, or establish whether data availability also influenced the start date. It therefore offers a concise rationale rather than a full examination of the sample design.

Key ideas

• Pre-1962 COMPUSTAT coverage is described as having serious selection bias.
• Earlier data reportedly favor large firms with successful histories.
• That bias can make older stock samples unrepresentative.
• The note does not rule out data availability as an additional factor.

Tags

# Answer by JorgeT (score 1)


# Why do Fama French's "Common risk factors in the returns on stocks and bonds" use data starting in 1963? Availability or convenience for results?












What is the reason Fama French's "Common risk factors in the returns on stocks and bonds" use data starting in 1963? Was it availability or convenience for results?

## Answer by JorgeT (score 1)

https://quant.stackexchange.com/a/37163

In their other paper from that year "The cross-section of expected stock returns" they mention that the COMPUSTAT data for earlier years have a serious selection bias; the pre-1962 data are tilted toward big historically successful firms."

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.