Lesson 4 of 7 · 13 min
Behavioural biases in analyst forecasts
Five biases (overconfidence, illusion of control, conservatism, representativeness and confirmation) distort forecasts, and each has a practical remedy built into the research process.
In short
- Overconfidence: too much faith in one's own forecasts; intervals that are far too narrow. Remedy: record and review forecasts, and use scenario analysis.
- Illusion of control: believing more data and more complex models will give control over outcomes. Remedy: model only regularly disclosed, material variables; talk to people with unique insight.
- Conservatism (anchoring and adjustment): updating too little when new information arrives. Remedy: regular team reviews, simpler and more flexible models.
- Representativeness, especially base-rate neglect: relying on the company-specific 'inside view' and ignoring the 'outside view' of the reference class. Remedy: start from the base rate, then adjust for real differences.
- Confirmation bias: seeking only information that supports the existing view. Remedy: seek out bearish analysts and colleagues with no stake in the idea.
Unlock this lesson free for 7 days
Create a free account to get 7 days of full access — every lesson, video, flashcard, mock and the question bank. No card needed.