Bitcoin Forum
March 19, 2024, 02:35:15 AM *
News: Latest Bitcoin Core release: 26.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 [47] 48 49 50 51 52 53 54 55 56 57 58 59 60 »
  Print  
Author Topic: Goomboo's Journal  (Read 281394 times)
oda.krell
Legendary
*
Offline Offline

Activity: 1470
Merit: 1007



View Profile
August 11, 2013, 01:57:29 PM
 #921

1d EMA22+21 with no threshold and 0.6 fees

2013-01-01 07:00 Simulation started. Balance: 1000 USD
2013-01-01 07:00 BUY 74.714 BTC at 13.30
2013-05-13 07:00 SELL 74.714 BTC at 114.82
2013-05-18 07:00 BUY 68.631 BTC at 123.50
2013-06-09 07:00 SELL 68.631 BTC at 107.89
2013-07-30 07:00 BUY 72.092 BTC at 101.48
2013-08-11 07:00 Simulation completed. Balance: 72.092 BTC (7425.52 USD)

Thanks.

So that's 72 btc vs. 74 btc that buy&hold would have given you. Supports the "the number are off" hypothesis.

I'll take a look at the code later (gekko is on github, right?) but I have an idea what could lead to the error, unless Erebusbat comes back earlier and solves the mistery :P

Not sure which Bitcoin wallet you should use? Get Electrum!
Electrum is an open-source lightweight client: fast, user friendly, and 100% secure.
Download the source or executables for Windows/OSX/Linux/Android from, and only from, the official Electrum homepage.
1710815715
Hero Member
*
Offline Offline

Posts: 1710815715

View Profile Personal Message (Offline)

Ignore
1710815715
Reply with quote  #2

1710815715
Report to moderator
1710815715
Hero Member
*
Offline Offline

Posts: 1710815715

View Profile Personal Message (Offline)

Ignore
1710815715
Reply with quote  #2

1710815715
Report to moderator
The Bitcoin software, network, and concept is called "Bitcoin" with a capitalized "B". Bitcoin currency units are called "bitcoins" with a lowercase "b" -- this is often abbreviated BTC.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1710815715
Hero Member
*
Offline Offline

Posts: 1710815715

View Profile Personal Message (Offline)

Ignore
1710815715
Reply with quote  #2

1710815715
Report to moderator
1710815715
Hero Member
*
Offline Offline

Posts: 1710815715

View Profile Personal Message (Offline)

Ignore
1710815715
Reply with quote  #2

1710815715
Report to moderator
S3052
Legendary
*
Offline Offline

Activity: 2100
Merit: 1000


View Profile
August 11, 2013, 02:30:24 PM
 #922

1d EMA22+21 with no threshold and 0.6 fees

2013-01-01 07:00 Simulation started. Balance: 1000 USD
2013-01-01 07:00 BUY 74.714 BTC at 13.30
2013-05-13 07:00 SELL 74.714 BTC at 114.82
2013-05-18 07:00 BUY 68.631 BTC at 123.50
2013-06-09 07:00 SELL 68.631 BTC at 107.89
2013-07-30 07:00 BUY 72.092 BTC at 101.48
2013-08-11 07:00 Simulation completed. Balance: 72.092 BTC (7425.52 USD)



if I read this well, this model does not outperform buy and hold: only +643%

our newsletter has fared better :-) (had to throw this in :-))

Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 11, 2013, 09:44:32 PM
Last edit: August 11, 2013, 09:59:10 PM by Marcus Antonius
 #923

I took a four month chunk off the end of the database to eliminate the abnormal spike in 2013, so the range is from June 2011 to January 2013. Buy and hold is -5%. Behold the results:

DAILY17181920212223242526
10344%335%317%331%327%323%311%311%330%324%
11335%346%331%347%328%311%326%332%324%329%
12348%332%347%324%316%362%324%329%336%341%
13347%339%324%341%362%324%340%334%382%394%
14339%324%334%368%324%327%337%382%433%370%
15324%363%359%329%327%337%419%431%366%376%
16363%359%329%330%337%419%422%369%376%415%
17
         
335%331%337%424%422%368%392%417%418%
18
         

         
337%424%410%368%419%417%418%362%
19
         

         

         
405%368%414%417%418%362%339%
20
         

         

         

         
414%417%437%362%339%339%
21
         

         

         

         

         
437%348%339%339%339%
22
         

         

         

         

         

         
339%339%339%292%
23
         

         

         

         

         

         

         
339%292%286%
24
         

         

         

         

         

         

         

         
286%285%
25
         

         

         

         

         

         

         

         

         
288%
TradesLikeAPotato
Newbie
*
Offline Offline

Activity: 41
Merit: 0


View Profile
August 11, 2013, 11:11:46 PM
Last edit: August 13, 2013, 04:20:25 AM by TradesLikeAPotato
 #924

I'm curious, is it best to use data from the conception of bitcoin or only data from recent months? Using all available data allows for testing under various market conditions, but using recent data gives you an idea what's most effective in current market conditions. I'm seeing conflicting reports on the most profitable EMA/time period, so I'm having trouble deciding the most profitable strategy.
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 14, 2013, 01:51:18 AM
 #925

More interesting backtest results. I compared Ichimoku, the 10/21 EMA, and the 15/24 EMA, looking at almost the full history of BTC trading, starting when one bitcoin went for a dollar, and ending today (1 BTC = $108.74). With the EMAs, I traded at the closing price of a candle which indicated a crossover, no threshold. With Ichimoku, I bought BTC when the price was over the Tenkan Sen over the Kijun Sen, and Chikou over price. I sold when the candle closed below the Kijun Sen. Trades were executed at the closing price.

Here are the results:

10/21 EMA P/L = 28,530%
Ichimoku P/L = 46,317%
15/24 EMA P/L = 50,028%

 Shocked

I share all this information gladly because I've learned so much from this forum.
StarenseN
Legendary
*
Offline Offline

Activity: 2478
Merit: 1362



View Profile
August 14, 2013, 09:29:21 AM
 #926

More interesting backtest results. I compared Ichimoku, the 10/21 EMA, and the 15/24 EMA, looking at almost the full history of BTC trading, starting when one bitcoin went for a dollar, and ending today (1 BTC = $108.74). With the EMAs, I traded at the closing price of a candle which indicated a crossover, no threshold. With Ichimoku, I bought BTC when the price was over the Tenkan Sen over the Kijun Sen, and Chikou over price. I sold when the candle closed below the Kijun Sen. Trades were executed at the closing price.

Here are the results:

10/21 EMA P/L = 28,530%
Ichimoku P/L = 46,317%
15/24 EMA P/L = 50,028%

 Shocked

I share all this information gladly because I've learned so much from this forum.

Thanks for sharing !

Which timeframe did you test for these results?
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 14, 2013, 11:14:58 AM
 #927

Daily.
Raffles
Newbie
*
Offline Offline

Activity: 10
Merit: 0


View Profile
August 14, 2013, 12:29:02 PM
 #928

Hi, I'm new here and just caught up on this thread! Great thread.

More interesting backtest results. I compared Ichimoku, the 10/21 EMA, and the 15/24 EMA, looking at almost the full history of BTC trading, starting when one bitcoin went for a dollar, and ending today (1 BTC = $108.74). With the EMAs, I traded at the closing price of a candle which indicated a crossover, no threshold. With Ichimoku, I bought BTC when the price was over the Tenkan Sen over the Kijun Sen, and Chikou over price. I sold when the candle closed below the Kijun Sen. Trades were executed at the closing price.

Here are the results:

10/21 EMA P/L = 28,530%
Ichimoku P/L = 46,317%
15/24 EMA P/L = 50,028%

 Shocked

I share all this information gladly because I've learned so much from this forum.

Wow those are startling percentages! Regarding 15/24, this makes me think of what Goomboo said about curve fitting. It is not surprising that this optimal combination for June 2011 to January 2013 also gives the best results up till now, because there is a big overlap in the data. What happens if you remove the overlap (i.e. only look at Feb 2013 to now)?
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 14, 2013, 06:09:11 PM
 #929

I don't know, dude, why don't you pitch in a little and tell us?
 Grin
TradesLikeAPotato
Newbie
*
Offline Offline

Activity: 41
Merit: 0


View Profile
August 14, 2013, 07:02:31 PM
 #930

That's strange. In my tests the 12/29 EMA has been most profitable.
Raffles
Newbie
*
Offline Offline

Activity: 10
Merit: 0


View Profile
August 14, 2013, 09:20:35 PM
 #931

So having spent a couple of days reading the thread, I've taken on board to have only 1% of your investment in a given fund and realised that I need to look beyond bitcoins alone to apply these techniques.

I've also taken on board the need to be selling short in down trends and that if you don't do that, you're halving the potential profit. So, I've been looking for places to invest that allow short selling and have a wide range of things to invest in. CFDs keep coming up. These seem ideal for a small time trader like me as there are no trading fees, however there seems to be an "overnight interest payment" or "overnight premium" associated with them if you hold at the end of the day. Is it possible to use CFDs for medium term trend following, or is it really for day trading only? How about sellling at the end of each day and then buying again at the start of the next to avoid the overnight interest - or is that a crazy idea?

One good thing, a lot of the sites have unlimited "simulated" trading on their platform, which is free.

Separately:
I don't know, dude, why don't you pitch in a little and tell us?
 Grin

Fair comment Smiley I just figured you already have it set up and only need to change the start date to get the answer, but maybe there is manual fiddling involved. Really appreciate you sharing what you already have. Thanks.
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 14, 2013, 09:44:04 PM
 #932

Fair comment Smiley I just figured you already have it set up and only need to change the start date to get the answer, but maybe there is manual fiddling involved. Really appreciate you sharing what you already have. Thanks.

I do it all manually.
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 15, 2013, 03:23:40 AM
 #933

That's strange. In my tests the 12/29 EMA has been most profitable.

What numbers do you get?
TradesLikeAPotato
Newbie
*
Offline Offline

Activity: 41
Merit: 0


View Profile
August 15, 2013, 05:42:35 AM
Last edit: August 15, 2013, 06:13:47 AM by TradesLikeAPotato
 #934

That's strange. In my tests the 12/29 EMA has been most profitable.

What numbers do you get?

12/29 daily gives 3456.193% profit, higher than any other EMA, using the time frame of the candles in this thread: https://bitcointalk.org/index.php?topic=239815.0
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 15, 2013, 11:19:45 AM
 #935

Quote
12/29 daily gives 3456.193% profit, higher than any other EMA, using the time frame of the candles in this thread: https://bitcointalk.org/index.php?topic=239815.0

That's not as high as the 20/21
TradesLikeAPotato
Newbie
*
Offline Offline

Activity: 41
Merit: 0


View Profile
August 15, 2013, 05:02:17 PM
 #936

Quote
12/29 daily gives 3456.193% profit, higher than any other EMA, using the time frame of the candles in this thread: https://bitcointalk.org/index.php?topic=239815.0

That's not as high as the 20/21

That's bizzare, 20/21 gives me 2312.392% profit on the same time frame.
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 15, 2013, 08:26:27 PM
 #937

With Gekko?
TradesLikeAPotato
Newbie
*
Offline Offline

Activity: 41
Merit: 0


View Profile
August 15, 2013, 08:55:52 PM
 #938

With Gekko?
Yes, all the default parameters other than EMAs.
Marcus Antonius
Newbie
*
Offline Offline

Activity: 57
Merit: 0


View Profile
August 16, 2013, 01:33:12 AM
 #939

I get 3415%. Odd. Something's amiss.
TradesLikeAPotato
Newbie
*
Offline Offline

Activity: 41
Merit: 0


View Profile
August 16, 2013, 02:31:40 AM
 #940

I get 3415%. Odd. Something's amiss.
Indeed. Just to verify, daily candles, 100 initial EMAs, .6 fee, and .25 threshold, correct? Anything you can think of that may be skewing the numbers?
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 [47] 48 49 50 51 52 53 54 55 56 57 58 59 60 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!