This paper might be of interest: "On the distribution of stock market data" http://lit.jinr.ru/Reports/annual-report05/ShortLITReports2004_StockMarket-163.pdf Quote: "[...] for most stocks the distribution of the closing prices normalized by traded volumes fits well the log-normal function. Do you see any practical implication?
Thanks for the paper. The log-normal distribution is the most accurate representation of the price behavior IMHO. There are, however, a few very simple methods to normalize this data to fit the normal curve. It is much more practical. But even without this conversion the log-normal character of market data opens up tremendous opportunities for the algo trading. Just think of the Arcsine law converted for log-normal distribution! Cheers, MAESTRO
The arcsin law adapted for leptokurtotic distribution of price changes, in layman terms, courtesy of yours truly: "Run your profits, and cut your losses short". Can anyone, except yours truly, put the law of the meander adapted for leptokurtotic distribution of price changes, in layman terms?.
That name was given to it by some homo that wanted to sound impontant several dozen years after I discovered it and named it!. Let's say it's a HOMOnym.
BlackâScholes show a practical application already. Also: http://en.wikipedia.org/wiki/Gibrat's_law
What's nice avout normalizing anythig? So u can compare apples to oranges and modelling is a lot easier.
What random walk, there ain't such a thing as random walk in trading. Cannot belief this thread is still going on, as my late father used to say: "wanking off in a dark suit may feel nice but I doubt anyone notices" Have a nice day