Factor Models - Risk or characteristics?
6 important questions on Factor Models - Risk or characteristics?
What two questions can we answer to determine whether something is a risk or a characteristic?
- Why doesn't the CAPM hold?
- Why do other factors predict returns?
Why might the CAPM not hold?
- Noisy beta estimation?
- What is the market factor (Roll's critique)
- Theory is wrong?
Why do other factors predict returns?
- Data mining (cynical view)
- Statistical result is coincidence, there is no real effect
- Risk factor (neo-classical view)
- Economic risk factor is causing expected return (if something is priced in the cross-sectional, then it must be a risk factor)
- Characteristic (behavioral view)
- Investors want to buy stock for other reasons
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
Why are policy makers interested in the characteristics?
- Policy makers: Mispricing -> mispricing implies inefficient capital allocation (capital flows to the wrong products), so suboptimal economic growth. If a factor is signaling mispricing -> ECB does somethng about it
Why are investors interested in risk or characteristic?
- Data mining -> do not invest (not real)
- Risk? -> invest if willing to take the risk
- Mispricing? -> Invest! Free lunch. Should eliminate mispricing
Can we disentangle the three (risk, characteristic, data mining) empirically?
- Data mining
- Should dissapear out-of-sample
- Risk
- Should not dissapear
- Characteristic
- Should dissapear after publication of the paper due to (risk) arbitrage
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding