Signal Processing - May 2017 - 24

Fundamental Analysis, Technical Analysis,
and Quantitative Analysis
Fundamental analysts analyze financial statements,
management and competitive advantages, competitors,
and markets to find the intrinsic value of a firm. Fundamental
analysis is the mainstream methodology for financial
professionals.
* Pick a company with mispriced intrinsic/correct
value
* Critique: subjective guesswork is required for future
growth and risks.
Technical analysts (chartists) analyze price and volume
patterns/charts to predict price movement and direction. Technical analysis is not mainstream among financial professionals.
* Buy low, sell high by analyzing price-chart patterns,
e.g., head-and-shoulder or double top/bottom reversal patterns.
* Technical indicators: moving average, relative
strength index, moving average convergence divergence, etc.
* Critique: the pattern definition is subjective and without rigorous statistical tests. Different chartists have
different opinions.
Quantitative analysts (quants) analyze financial markets using complex mathematical and statistical modeling, measurement, and research of historical data.
* All quantitative risk modeling and algorithm trading,
and increasingly more hedge funds, are based on
quantitative analysis.
* Critique: many quantitative analyses lack economic
justifications. History may not represent the future.
The validity of the analysis is subject to various
model and data mining risks.

beyond the Cramér-Rao bound, and 3) statistical tests against
alternative models, because the confidence in the model used
is established from the understanding of physical laws rather
than purely from data. For example, to track a flying missile,
we have basic understanding of its motion model based on
Newton's laws and sources of noise. However, we do not have
such understanding of the market price movement other than
common sense. The statistical test is often the only quantitative
tool on which we can rely.
Thus, for stock price prediction, researchers need to ask
themselves, Are we confident that we have done enough
tests to bet the money in the market? Are we confident that
our price model beats the mainstream asset-pricing models
and the collective information processing (crowdsourcing)
of all other market participants? Note that market equilibrium is reached by competing arbitrage-seeking efforts in the
market. Apparently, there is a large burden of proof against
the EMH.
24

Prospect Theory: Value Function
Value

Losses US$1,000 Loss

Gains
US$1,000 Win

FIGURE 4. The prospect theory assumes that people make choices according to the value function. The amount of pain felt from losing US$1,000 is
about double the amount of joy felt from winning the same amount.

Prospect theory and behavioral economics
Both the expected utility theory and the EMH assume that
people, or at least some people, act rationally and optimally to
exploit risk-free profit in the market to maximize the expected
utility. Therefore, the only source of excess return is risk. The
expected utility theory and the EMH were dominant before
the 1980s in academia.
However, people frequently argue that it is often the irrational or emotional decisions that drive the market, generating bubbles and recessions. Such psychological overreaction
or underreaction to information moves the market, generating mispricing and predictable market movement patterns,
and thus possible risk-free profitable opportunities. This
line of thinking has led to the development of behavioral
economics, embraced by many people eager to justify price
anomalies and patterns, both academics and (more) practitioners, as it provides a formidable (desirable) alternative to
the EMH.
In 1979, the psychologists Daniel Kahneman and Amos
Tversky developed prospect theory [7], which has become a
foundational theory for behavioral economics. Daniel Kahneman won the 2002 Nobel Memorial Prize in Economic Sciences for his foundational work in behavioral economics (Amos
Tversky died in 1996).
In contrast with the expected utility theory, prospect theory
states that people make choices based on the value function
illustrated in Figure 4. The value function states that people
make financial decisions not from the expected utility of the
absolute total wealth but from the psychological and emotional
perceptions of joy and pain depending on the relative gain or
loss. The function is asymmetric in that the amount of pain
from losing US$1,000 is about double the amount of joy of
winning the same amount. The gain part is concave, while the
loss part is convex. A loss-aversion gambling example is shown
in "Take a Bet Again-Intolerance of Loss?"

IEEE Signal Processing Magazine

May 2017

Table of Contents for the Digital Edition of Signal Processing - May 2017