Business & Strategy |
Equity Investments: Hot Stocks |
BANKING & FINANCIAL INDUSTRY |
5 more ...|
|
||
|
Activity:
2 comments
334 views
last activity : 07 06 2010 20:18:04 +0000
|
||
|
|
Bell Curve VS Poisson Curve, implication on Modern Finance
The history of Bell curve is that, the French Emperor Louis V wanted to calculate the location of planets accurately when there, was no such methods available then. He contacted Legendre the Mathematician. for this and Legendre recommended Gauss to the emperor. Gauss started the work measuring the orbits, and predicted their location using the Normal Bell curve which he discovered for this purpose. He figured out the standard deviation and this could predict the location of all planets, if one new the location of any one of them. The standard deviation from the mean for each predicts the location for any time. The history of Bell curve is described here for the reason that it was originally invented to predict the events that are consistent, like planet locations that don’t change with time but are related to one another at any point in time, so to say they are constant.
Now Poisson created his curve, to predict the death of atomic particles, where the time series formed are not dependent on the previous data.
Now, the Social Scientists comes along and starts applying Normal curve to predict events whose time series data in most cases are not related to each other(each data time series, not the cross sectional data) through correlation, which is a serious error. Modern finance theory uses it as if they are not aware of this fact.
Take for example Black and Scholes formula, where one of the preconditions for it to work it seems is the existence of an Efficient Market. Efficient Market stipulates that one days data is not in any way related to the next days data. B&S then uses the standard deviation sigma to calculate risk which is based on Normal Bell curve which is for data that are related. I live it to you to ascertain the consequences.
Again Capital Asset Pricing Model stipulates efficient market precondition, and beta in it is calculated using Normal curve assumption to data that are not correlated partially, which means again the value may be erroneous or may not be stable enough for long.
It has become a practice where a Professor seems to transfer some error to the student and it goes on for years they work like a tag team one complementing the other and spawning theories that correlate in some means.
Bell curves works better in Physical Phenomenon and is quite reliable in Physical and Health Sciences, rather cannot be depended on when used in Social Sciences. In Physical and Health Sciences the Normal curve works because the averages of the data depicting the phenomenon always are sufficient to generate the required functionality where as Social Sciences of Markets the averages of data don’t work since Humans are discerning by nature and they are capable of price discovery which may be very different from expectations and deviations and can bring about foul outcomes.
Company forecasts based on Bell curves seems to be. accurate since cetris-paribus states of the economy and company missions and strategies makes their individual data more dependent on previous years.
|
|
|
|
|
|
|
|
|
|
|
|
All relegious processes releases the liety from 'unconscious' mental state to 'conscious' mental state. Even 'possessions' are similar when one is possessed one is totally unconscious and when he is worked on by a wizard he uses sticks etc; which in... |
It is part of a humans development stage during child hood when it happens during later age it means the development of the person is not in tandem with each stage it simply got prolonged to later times. Also it can be called a 'transference neurosis'... |
Communication is very important in business management, then when everyone is a neurotic out there to cover up their mistakes then communication fails and chaos ensues. Thanks for the referral. |
If the occurrence of an event is random and rare, then the expected distribution function for this event is expected to follow a Poisson distribution. The Poisson distribution describes a number of discrete events in a sequence. Some nice properties of the Poisson processes is that the data variance equals the expected value: oar(x)=E(x) (Often seen in storm and hurricane events). The events $\tau_i - \tau_{i-1}$ are ``exponentially'' distributed if the events are of a Poisson type. The transform $\sqrt{x}$ is more Gaussian. The Poisson processes assume that the probability of events is small to avoid having multiple events at same time. The expected number of occurrence only depends on the length of the interval over which they are counted: the occurrence does not depend on time