**Behavioral finance FAQ / Glossary (Quant - Quantitative)**

This is a separate page of theP-Qsection of the Glossary

Dates of related message(s)in theBehavioral-Finance group(*):

Year/month, d: developed / discussed,

i: incidental

Quant01/12i - 02/01,7i - 05/2i -07/9i

+ see quantitative analysis /

investment, stochastic, model

Heard in The Stochastic Café, on Greek Avenue:

"My equation is better than your equation"

Definition:

Quant (quantitative analyst) is a colloquial way (*) to call a

financial practitioner who applies

advanced mathematics (**)to market data, for

and computing software

analysis, simulation and decision making.

(*) "Rocket scientist" is another popular appellation.

(**) Those mathematics cross a multitude of data ("big data"),

from fundamental ratiosto marketfluctuations and try to

findstatistical correlationsand toapplystochastic(see stochastic) based on the

calculations

probabilities of future events.

History and applications

The golden years of Elvis and Markowitz

From the electric guitar and the physics lab

to the market pit.

There is a before and an after, here!

Until the first half (included) of the 20th century,

asset portfolioapart using some

management,economic and financial analyses

(and private info), was the realm ofrules of thumb(some of them

quite sound), spiced withactuary mathematics(quite relevant

also) and someprice chart observations(supposed to show a

path through the market minefields).But in the 50's - 70's, not only Rock and Roll, but also

physicalmaths(QA), took over the stage.

applied to financial quantitative analysis

Quantitative Analysis started with the first academic

works that launched the"modern portfolio theory /.

modern financial theory"

It includes for example the CAPM (see that acronym)

and the Theory of options.They make an

intensive use of stochastic calculation(see that phrase).

Random laws served for breakfast, lunch and dinner.The main applications were, and still are:

Quantitative value analysis:see the related article

below,

Quantitative investment decision

models:see model, quantitative investment, system trading.Quants work mostly in banks and hedge funds and deal mostly with

operations that involve derivative contracts.

Rationale and limits

Can equations cover all real life cases?

Or even avoid a piano to fall on you from the sky?However rational and useful are the methods used by quants.

They bring the risk of an

excessive trustin

statistical series, mathematical formulas (see

quantitative analysis, numeracy bias) and random

laws (see anomalies).The advantage is to avoid human biases. One of

them is the "base rate neglect" (see that phrase) that

shuns probabilities, although they are crucial tools,

at least in standard / well known situations .

The main flaw might be that quants tend to

the probability / frequency of

underestimate

extremerisks("black swans"), when the situation

leaves its usual / standard track.

In other words, they might confuse (well known)

risk, and uncertainty.

What is deemed to happen only once in a billion years

according to the normal law of distribution (see distribution, rare

events...) might in fact have a real probability to occur

...in the next five years.

As most QA tools ignore or understate the possibility of events at

the same timeexceptionaland disastrous,

the consequences might be dramatic.

Those tools usually do not prepare for those

rareevents,and are useless (or even

counterproductive) to help manage them.Human imagination, historical research and narrative methods,

when not biased, might be superior to standard math in defining

(and dealing with) someextreme scenarios.Specifically

those in which market price volatility changes its,

naturefrom an ordinary vibration reflecting a classical law

of randomness,

to something more chaotic(seedynamical).

systems

A typical example: illiquidityMost quant based trading models consider counterparts to be

always available, therefore allowing to hedge and make arbitrages

whenever needed.Therefore, they become

impotent when a liquidity crisis (see

that phrase) strikes.

Probability lawsused in normal times become obsolete in the

middle of a chaos and crisis akin to the 100 year storm.

The market models that quants use

The"quantitative analysis"article below gives details about the types of

models that quants like to use.Those tools make a massive use of mathematics, based largely on

stochastic(a subfield of

calculationprobabilitiesthat applies to randomdynamical: see the phrase).

systems

The risk-return parameters which are used (beta, gamma, delta...)

are called colloquially"the Greeks".

Quant fundSee quantitative investment

Quantitative analysis / QA01/10i + see models, numeracy,

probability, model, stochastic

Questioning the market with a computer chip.

Quantitative (investment) analysis / QA

covers two kinds of practices

(that coexist or not in the same model):

1) Asset value

calculations

Those equations / models take into

account various data and parameters

related to:The company (

endogenousdata)The market (

exogenousdata):

expected incomes, interest rates, risk

premia, volatility...

2) Investment /

tradingmodels.Here we go from

valuation to money

management

decisions

(see "quantitative

investment").

Those models try for example to detect

"buy orsellsignals"and "arbitrage

opportunities" when asset prices and

returns deviate from "normal" random

distribution laws.A model's

time horizonis important,

as a robot with a short time view might

misunderstand what signals those

deviations give, not seeing that a blip

might show (or hide) the start of a

trend.Many of those models are based on

"stochastic calculation"

(see that word).It is a field of probability-related advanced

mathss that applies todynamical systems that follow random

processes(this is not the case at all times for all dynamical systems - see that

phrase - as many of them, here is the trap, often diverge from

standard probability laws).Also, among the investment models, some are akin to technical analysis,

but in a more sophisticated and mathematical way that try

Either to spot a

short term "anomaly"that opens a possibility of

fast reversion to the statistical average or to the middle of the trend

corridor.Or, on the contrary, the

start of a new trend, thus opening the

opportunity of a new medium term strategy.

Does QA opposes or complements

other market approaches?

Still not the do-it-all / Swiss army knife of finance.

QA is one of the four analysis tools that market professionals usually quote :

BA, FA, QA, TA (see those acronyms).There is still no "string theory" to unite those types of analyses.

It is true that they have at least as many points of contradictions than of

convergence.This tool might detect recurrent market anomalies, by comparing its

findings to what an "efficient" model would give.It can thus be

used as well in everyday trading as in behavioral.

finance research

That is what makes

quantitative behavioral finance(see

that phrase) one of the behavioral finance sub branches.

? or ?

Quantitative analysis and their models

* also have

* have their usefulness as seen above.drawbacks, as seen in the

"model" article and summed up below :

They might fall in the

"numeracy" biastrap (see that phrase), asin

an excessive trustPast series,

Theoretical equations,

Standard random laws,

Market indicators such as "implied volatility"...,

They might become

used blindly, as standard recipes in

cases to which they can not be applied soundly, therefore turning

into a bounded heuristic and a herding tool,

They are at a loss to foresee and/or take into account some

eventsunknownto themodel.This is the case of:

* events or combinations of events

without known precedent*

"rare events"(see that phrase), for example suddenliquidity

crises(see liquidity squeeze).

Robots' fuses blow when exceptional events strike!A broader economic and market culture and experience might be

able to compensate somewhat that overconfidence in mechanical

prevision tools.To conclude on a more general psychological topic, too much trust in

directlycan lead to the well knownavailable or well known data

paradox, of looking for one's lost keys in the street at night, not where

they might have fell, but under the lamppost because it is where there is

enough light.

Dates of related message(s)in the

Behavioral-Finance group (*):

Year/month, d: developed / discussed,

i: incidental

Quantitative behavioral financeSee behavioral finance

Quantitative investment / trading

01/12d + see systems,

system trading, model,

algorithmic trading

Robotic money dealersQuantitative investment / trading is the use of

mathematical models(see

"model", and "quantitative analysis"):

To make previsions of

returns(due mainly to price

moves),To evaluate

risks,

To help

decidefinancial portfolio operations (arbitrages),so as to optimize

asset allocationordiversification,Mostly for day-to-day

trading(or even "system trading", see

that phrase).Therefore less for long term investing (except if the model

includes it).This short term inclination, not to say bias, contributes

to an abundant use of sophisticatedderivative contracts

devised by ...quants.Those investment models take thus into account, among other parameters,

and related data.

risk-return ratioThey are used by - among other financial institutions - some mutual funds

called"quant funds".Also some commercial software is available to individual traders.

Quantum jump, lump00/6i,10i,12i + see percolation

threshold, Markovian jump

When a market suddenly chooses to ride a rocket

or to dive from a cliff.The quantum word is taken here just as a metaphor.

It means that market prices and returns might

leap suddenly from

one behavioral phase to another.Such an abrupt change could be linked to percolation (see that word),

Markovian jumps, external shocks...

(*)To find those messages: reach thatBF groupand, once there,

1) click "messages", 2) enter your query in "search archives".

BF pollsMembers of the Behavioral Finance Group,

please vote on the glossary quality at

This page last update: 14/07/15

P-Q section of the Glossary

Behavioral-Finance Gallery main page