Introduction of Brownian Motion
Introduction of Brownian Motion
Introduction of Brownian Motion
MOTION
NAME LEE MUN YEE CHAI SHUK YNG LEONG LEE SHIN LEE WAN YEE MATRIX NUMBER UK20119 UK20160 UK20323 UK20355
INTRODUCTION OF BROWNIAN MOTION Definition Of Brownian Motion A real-valued stochastic process W() is called a Brownian Motion or Wiener process if (i) (ii) (iii) W(0) = 0 a.s., W(t) W(s) is N (0, t s) for all t s 0, for all times 0 < t1 < t2 < < tn , the random variables W(t1),W(t2) W(t1), . . . , W(tn) W(tn-1) are independent (independent increments). Notice in particular that E (W (t)) = 0 and E (W2 (t)) = t for each time t 0.
The Central Limit Theorem provides some further motivation for our definition of Brownian motion, since we can expect that any suitably scaled sum of independent, random disturbances affecting the position of a moving particle will result in a Gaussian distribution. Motivation of Brownian Motion In 1900 L. Bachelier attempted to describe fluctuations in stock prices mathematically and essentially discovered first certain results later rederived and extended by A. Einstein in 1905. Einstein studied the Brownian phenomena this way. Let us consider a long, thin tube filled with clear water, into which we inject at time = 0 a unit amount of ink, at the location . Now let denote the density of ink particles at position and time . Initially we have , the unit mass at 0. Next, suppose that the probability density of the event that an ink particle moves from to in (small) time is . Then (1) But since is a probability density, ; whereas , the by
Sending now
we discover
This is the diffusion equation, also known as the heat equation. This partial differential equation, with the initial condition
This equation and the observed properties of Brownian motion allowed J.Perrin to compute ( = number of molecules in a mole) and help to confirm the
Following
note
from:
http://www.doc.ic.ac.uk/~nd/surprise_95/journal/vol1/skh1/article1.html Geometric Brownian Motion Assumptions GBM describe the probability distribution of the future price of a stock. The basic assumption of the model is as follow: The return on a stock price between now and some very short time later (T-t) is normally distributed The standard deviation of tis distribution can be estimated from historical data ST volatility is good predictor of the LT volatility
The stock price models employed in option pricing are not predictive but probabilistic: they assume a distribution of the future prices derived from historical data and current market conditions. Geometric Brownian Motion Mean and Standard Deviation The mean of the distribution is times the amount of time (The expected
It says that short term returns alone are not a good predictor of long term returns. Volatility tends to depress the expected returns below what the short term returns suggest (the average amount the stochastic component depresses returns in a single move is )
The standard deviation of return increases in proportion to the square root of the amount of time (Bachelier, 1900)
APPLICATIONS OF BROWNIAN MOTION TO MARKET ANALYSIS Brownian Motion in the Stock Market In the middle of this century, work done by M.F.M Osborne showed that the logarithms of common-stock prices, and the value of money, can be regarded as an ensemble of decisions in statistical equilibrium, and that this ensemble of logarithms of prices, each varying with time, has a close analogy with the ensemble of coordinates of a large number of molecules. Using a probability distribution function and the prices of the same random stock choice at random times, he was able to derive a steady state distribution function, which is precisely the probability distribution for a particle in Brownian motion. A similar distribution holds for the value of money, measured approximately by stock market indices. Sufficient, but not necessary
conditions to derive this distribution quantitatively are given by the conditions of trading, and the Weber-Fechner law. (The Weber-Fechner law states that equal ratios of physical stimulus, for example, sound frequency in vibrations/sec, correspond to equal intervals of subjective sensation, such as pitch. The value of a subjective sensation, like absolute position in physical space, is not measurable, but changes or differences in sensation are, since by experiment they can be equated, and reproduced, thus fulfilling the criteria of measurability). A consequence of the distribution function is that the expectation values for price itself increases, with increasing time intervals 't', at a rate of 3 to 5 percent per year, with increasing fluctuation, or dispersion, of Price. This secular increase has nothing to do with long-term inflation, or the growth of assets in a capitalistic economy, since the expected reciprocal of price, or number of shares purchasable in the future, per dollar, increases with time in an identical fashion. Thus, it was shown in his paper that prices in the market did vary in a similar fashion to molecules in Brownian motion. In another paper presented around the same period, it was also found that there is definite evidence of periodic in time structure (of the prices in Brownian motion) corresponding to intervals of a day, week, quarter and year : these being simply the cycles of human attention span. With compounding evidence and widespread acceptance that Brownian motion exists in market structures, many researches and studies have since taken place, revolving and evolving around this theory. For example, a statistical analysis of the New York Stock Exchange composite index to show that Levy Processes do exist in it was carried out by R.N.Mantegna, and he showed that the daily variations of the of the price index are distributed on a 'Levy' stable probability distribution, and that the spectral density of the price index is close to one expected for a Brownian motion. Investment, Uncertainty, and Price Stabilization Schemes Another application of Einstein's theory is seen in the paper done by William Smith, who uses the method of regulated Brownian motion to analyse the effects of price stabilization schemes on investment when demand is uncertain. He investigates the behaviour of investment when price is random, but subject to an exogenous ceiling,
and with the aid of the mathematics of regulated Brownian motion, demonstrated that price controls mitigate the response of investment to changes in price, even when controls are not binding. The conclusions developed would be applicable to any economic situation involving smooth costs of adjustments of stocks when prices are uncertain but subject to government control (ie. rent controls, hiring/firing decisions in the presence of a minimum wage). A Brownian Motion Model for Decision Making The Brownian model was also made use of by L.Romanow to develop a model for a decision making process in which action is taken when a threshold criterion level is reached. The model was developed with reference to career mobility, and it provides an explanation of an important feature of promotion processes in internal labour markets. The model assumes continuous observation of behavior (of employees) and that the only route for leaving a job is by promotion. This suggests that the important mechanisms in the process are the basic evaluation procedure -- rating which includes a random component (Brownian motion theory), and the decision rule -- promote when an estimated average reaches a criterion level. The model was able to provide substantive qualitative results and hence is of good use to the 'real' world in decision making policies. Stock Prices and Its Relation with Brownian Motion Prices changes can be decomposed into two components: Deterministic component (the guaranteed 10 percent compounding each year) Stochastic or random component (the plus or minus 5 percent jump experienced each year on top of the guaranteed 10%) The stochastic component is normally distributed with an expected value of zero (is symmetric about zero, just as the random jump was symmetric about zero) The standard deviation of the stochastic component controls how much volatility there is on top of the deterministic component. The long term returns on a stock are proportional to
Because X2 itself represent the result of two price moves, the average amount the stochastic component depresses returns in a single move is .
If a random variable representing the stochastic component of Brownian Motion, then we have
For example: = 10% per annum = 35% per annum The model predicts that the five-year returns are normally distributed, with mean and standard deviation .
If we observed the average one-day returns and the standard deviation of one-day returns, respectively, we would find that they are approximately and
Geometric Brownian Motion and The Real World Geometric brownian motion (GBM) states that the mean and standard deviation of a stock are constant. Clearly this is not the case with the rate of return on a stock (in fact, the rate of return cannot observe directly). Fortunately,only the instantaneous standard deviation is important for option pricing and standard deviation is less difficult to predict if we compare the change in the stock price over small period of time, infinitesimally close.
is the volatility parameter, growing at a rate of the square root of time, and is a simulated variable, usually following a normal distribution with a mean of zero and a variance of one. Now, we calibrate the model by computing its parameters over very short time intervals, and then using the conclusions to infer information about the long-term returns and volatility. Geometric Brownian Motion As A Tool for Studying The Stock Markets
Geometric Brownian Motion- Empirical Evidence Geometric brownian motion states stock returns are normally distributed, and should be proportional to elapsed time and standard deviation should be proportional to the square root of elapsed time. What does the data say?
Large movements in stock prices are more likely than GBM predicts (the likelihood of returns near the mean and of large returns is greater than GBM predicts, while other returns tend to be less likely).
Downward jumps three standard deviations from the mean is three times more likely than a normal distribution would predict (the theory underestimates the likelihood of large downward jumps).
Monthly and quarterly volatilities are higher than annual volatility and daily volatilities are lower than annual volatilities.
The Distribution of Stock Prices GBM model concludes that stock returns are normally distributed and the stock prices are lognormally distributed The annualized return is given by The above equation is equal to Lets write x for this random variable Rearranging a bit, we have a new random variable: X ( the return, a random variable, normally distributed) plus a constant (not random) Therefore, the natural logarithm of the future stock price ( distributed ) is normally