Stock Price Modeling Based on Stock Prices Following Geometric Brownian Motion
Long-term returns of stock market should be greater than the yields of long-term
bonds, because otherwise, no one would buy stocks, causing stock prices to fall, price-to-earning ratios of stocks to rise. In the meantime, Everyone would just buy
long-term bonds, causing bonds' prices to rise and their yields drop.
The 5% shown in the above table can fluctuate and follows a normal distribution,
and according to the definition of log-normal distribution: "the logarithm
of random variables of a log-normal distribution are normally distributed", the
stock prices follow a log-normal distribution.
Mean of monthly return:
Since stocks are thought to be riskier than bonds, there is a "risk premium"
associating with the risk.
For example, we the 30-year bond yield is 2%, stock market return should be
greater than 2%, let's say 7%, the 5% difference is the "risk premium".
This 7% can be viewed as the mean of stock market return.
Geometric Brownian Motion
Brownian motion means constant random movements. Stock price "return rates" follows Brownian motion which means stock prices follow a geometric Brownian motion.
Fluctuation
Stock prices fluctuate. No one knows the short-term movement of stock prices.
One way to guess how they move is to use the normal distribution to model the
"returns" of stock market.
If the return of a stock a normally distributed, its price is normally
distributed.
| Time | Price | ln(Price) | Return |
|---|---|---|---|
| 1 | 100 | ln(100) | 5% |
| 2 | 100e5% = 105.13 | ln(100) + 5% | NA |
Modelling
We model the stock price movement by using the javascript "Math.rendom()"
method.
Suggested reading: Application of the central limit theorem: modeling normal distribution using javascript
%
Standard deviation of monthly return:
%
Time:
month(s)
Comments
Post a Comment