Factor Model and APT Flashcards
The Arbitrage Pricing Theory (APT)
uses no-arbitrage conditions as a basis for determining the structure of asset
returns.
Factor models
are statistical models that can do this. They use actual portfolios whose return is unambiguous and easy to measure.
Factor models assume that asset returns depend on some common factors.
A single factor model assumes
llassetreturns depend on the systematic risk of one common factor.
When the common factor is the market index it is called
the single index model (SIM). The SIM is the most common single factor model.
Sim uses realised excess returns
Comparing SIM to CAPM the ALPHA would
A positive alpha (αi > 0) would imply an asset’s return is higher than predicted by CAPM. A negative alpha (αi < 0) implies the asset’s return is lower than predicted by CAPM.
Multi-factor models
extend the SIM to include more than one factor.
he factors need to be a source of systematic risk and have their own risk premium. Some examples are:
• Excess returns on the market portfolio
• Unanticipated inflation
• ect
A Two-factor Model assumes?
Thee are two systematic factors, a market portfolio and a treasury bond portfolio
ARBITRAGE PRICING THEORY (APT)
Assets are mispriced if they do not have a zero alpha. Arbitrage trading will exploit mispricing by buying an underpriced asset financed by short selling an overpriced asset.
The APT assumes
– Markets exploit ‘arbitrage’ opportunities.
• This forces alphas to zero, and makes expected returns proportional
to betas (sensitivities to risky factors).
– Asset (portfolio) returns are determined by several risky factors.
• However, it does not specify what the factors should be.
– A sufficiently large number of assets exist so that idiosyncratic risk can be diversified away.
• This only works on a large well diversified portfolio.
