Efficient Markets Flashcards
(20 cards)
When is a market efficient?
A market is efficient if prices Pt include all relevant information up to t.
Weak form efficiency
Actual prices fully reflect all information contained in past prices.
Looking at past price trends to extrapolate current prices.
Semi-strong from efficiency
Actual prices fully reflect all public information, like fundamental analysis of companies/countries.
Economic and geopolitical environment, sector and stock specific information.
Any public information is factored in and forms the price.
Strong form efficiency
Actual prices fully reflect all informaton, including private information.
Predicting prices…
If markets are efficient, people can predict price changes over a period given the information they access.
The change in prices is a positive drift on top of which, there is noise (unpredictable news).
A change Pt - Pt_1 is only due to ‘news’ arrived between t-1 and t, so returns cannot be higher than a fair payment for undertaking risk.
How does the efficient markets hypothesis deal with CAPM and other factor pricing models?
They are fully consistent with market efficiency.
These models uncover sources of risk, and through the beta of each factor, they link the factor risk to the excess return of an asset.
Key assumption for market efficiency:
If there is serially correlated information, this is noticed and acted upon.
Past information helps inform future prices.
How can inefficiencies be exploited?
If an individual has information that others don’t.
Can efficient markets have lagged prices?
Price lags cannot be statistically significant.
They should not inform returns in a statistically meaningful way.
In efficient markets, can knowing past prices give a predictive edge?
No.
Asset prices are a random walk (price changes are unpredictable).
How can we model weak form efficiency?
Price processes are martingales.
No patterns in past prices can help predict future returns.
No drift in predictable returns, changes are purely due to new information.
Martingale Condition
Unpredictability of assets’ returns is expressed by the martingale condition.
Martingales express the notion of a fair game, which is neither in your favour or your opponent’s.
A fair game must be independent of history and can only depend on the last realisation.
But with historical positive returns, there is a drift.
If a geometric random walk involves discrete time, what happens when time intervals are infinitesimally small?
With infinitesimally small time intervals, the process converges to a Geometric Brownian Model.
What is the fundamental price of an asset.
In a risk-neutral world, an asset price is only determined by its mean return μ given by the drift of its stochastic process.
This is the forward value of an asset price.
How to forecast the drift μ?
The return (or drift) μ is linked to the fundamental price of an asset.
Rational expectations are the forecast of this mean μ.
Stock prices reflect business activitiy with the purpose of generating profits such that there is a positive rate of growth.
Rational Expectations
Given an information set of all relevant asset information, rational expectations are the mean price obtained from expectations formed based on information given.
Rational expectations are a limit: agents select forecasts that minimise the forecast error, with the crucial assumption of a zero mean error.
And errors are uncorrelated.
If people form expectations in a way that tries to exploit correlation, they are not rational.
Why does The Martingale look to be a necessary condition for an efficient market?
An efficient market is one in which the information contained in past prices is instantly, fully and perpetually reflected in the asset’s current price.
Efficient markets are then equated to the existence of a Martingale.
Non-overlapping price changes are uncorrelated at all lead and lags, which implies the ineffectiveness of all linear forecasting rules.
What if the rational expectation theory is wrong?
Error models, feedback, correlated expectations.
If it is wrong, an “error model” can be constructed to exploit market inefficiency.
The interaction of markets’ participants behaviours can lead to instabilities due to rapid positive (self-reinforcing) feedback.
Expectations can become correlated, leading to a growth of volatility, instabilities, bubbles and chaotic dynamics.
Operational definition of market efficiency:
Financial markets are efficient if no one can consistently earn excess returns… i.e.,
* After risk is factored in
* Costs are factored in (transaction and analysis costs)
What should be true if markets are efficient?
Security prices should respond quickly to new information.
Professional investors should not outperform net of fees.
Simulated trading strategies should fail.
