Recent content by tdazio

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2956161

this page can help you: http://datagrid.lbl.gov/backtest/index.php

Euan Sinclair suggests to keep things simple: http://blog.factorwave.com/volatility-forecasting-for-traders

1896, DJI http://stooq.com/q/d/?s=%5Edji 1789, SPX http://stooq.com/q/d/?s=^spx 1709, UK index http://www.bankofengland.co.uk/research/Pages/onebank/threecenturies.aspx

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2745220

Interesting question. But if price raises from 100 to 150 (typical elite-syle backtest profit :-) log-ret error became intollerable To avoid error in presence of big returns (positive or negative), I agree with benjiamin m gross recipe here...

Are you sure? Averaging daily log-return to get yearly return introduce error: it seems me need trasform numerator of SR formula to a linear geometric return..

Daal, kelly f is: (r-Rf)/v, were r = expected return, Rf= free-risk ret, v= expected variance of ret Tipically you use history to forecast r and v. If r and v come from past real trading 1/2 kelly is ok, but if they come from backtesting I suggest 1/4 kelly or less. Better to have not max...

cjbuckley, E. Thorp in his famous 'The Kelly criterion in blackjack, sports betting and the stock market' (2006), shows that Warren Buffet is (was?) essentially a full kelly bettor. But probably Mr. Buffet is not real-world, is an other planet :-)

in real-world, where average sharpe-ratio are in order of 0.5, continuous finance kelly criterion (f=1) works great. of course, in dream-world of data-snooper backtester kelly fails, but the fault, dear Brutus, is non in our stars , but in ourselves :)

in R: alarm() or using package tuneR : http://music.informatics.indiana.edu/courses/I546/tuneR_play.pdf

The core of strategy (long overnigth, short intraday w/o filter) is here (credit to Jay Kaeppel): http://www.optionetics.com/market/articles/2012/11/28/kaeppels-corner-the-greatest-gold-stock-system-youll-probably-never-use Since November 2012 (date of Kaeppel's post), out of sample and using...

Yes, but only using past tense verbs

Take GC futures 5 minutes data stream: #1 at 09:30NY record price P1 #2 at 15:40NY record price P2 and #3 if P2 > P1 then: #3a long overnite GDX using MOC (entry)/MOO (exit) order #3b next day short intraday GDX using MOO/MOC order (or long DUST) #4 if P2 < P1 go to sleep (or to fish, or...