Ninjastic
PostEdited versionsQuotes to this post
Post
Topic
Board Services
Re: B-BOT: New MtGox EMA Trading Plataform
by
fible1
on 12/05/2013, 17:46:05 UTC
Quote from: eduk on May 12, 2013, 05:40:19 PM
Quote from: fible1 on May 12, 2013, 05:34:56 PM
Quote from: IdiotHole on May 12, 2013, 05:09:02 PM
How about some sort of ranging market detection so that the bot would stop trading in periods of low volatility like we've seen over the last few days?

The link in this post seems very interesting: https://bitcointalk.org/index.php?topic=67591.msg2118599#msg2118599

Hi Smiley,
  Yes I saw the post and I've been thinking about it. An EMA trading bot, by its very nature (it identifies trends), is resistant to volatility if proper time thresholds are used (15 minutes to an hour in the BTC market), you can see a full discussion here: https://bitcointalk.org/index.php?topic=60501. The solution on the thread is more designed for a frequency bot.

What this solution is doing is its taking the lowest value from the current set of data and subtracting it from the highest, then comparing that value to the last value obtained from the last set of data. It's an unreliable indicator at best.

The solution implemented here is very easy for us to add to the bot and I'm happy to schedule it, I have yet to turn down a request for the new version Smiley, but please be aware it is a highly unstable solution that will likely be counterproductive. I leave you with the words of the programmer who added the feature in the bot you are discussing:

"Whether this is the best way of working out volatility, I'm not sure." - Eduk

What I can promise you is that we will be sure to publish a complete and effective volatility solution when we have found one and tested it out with our own money Smiley.

Pablo.



I'm not sure because I have very little trading experience and there is most likely better ways of working out volatility. In this very recent example which I'm currently trading with, it skipped the EMA thresholds because volatility wasn't high enough and avoided a stupid sell and buy based on volatility.



Eduk, I can see that, nice job!

We will look into implementing our own solution once we have had time to design one and test it out. Please let us know how your solution works out in the next couple of weeks, if you've found something we missed, we would love to collaborate; and perhaps our own solution will be effective for your bot as well Smiley.

Best,
   Pablo.