Im looking for a framework/library to program a trading bot. Ideally Python but java or even C would be acceptable. Anyone can recommend any?
Just to be clear, Im not looking for an already programmed bot or strategies. Im looking for a library that it will facilitate writting my own.
I've got a python system the grabs the Gox and CampBX feeds and stores it to a postgresSQL database. And I've written some simple queries across the data... |
|
|
|
Gold down today, bitcoin UP  the silverbox update (comparison from the beginning of this thread, March 13th, 2012, gold=1690, Bitcoin=5.4): Bitcoin is 95.70. Gold is 1598.10 Bitcoin: 1672.22% Gold: -5.44% Diff: 1774% advantage Bitcoin and Growing
|
|
|
|
|
the silverbox update (comparison from the beginning of this thread, March 13th, 2012, gold=1690, Bitcoin=5.4):
Bitcoin is 89.50. Gold is 1605.80
Bitcoin: 1557.41%
Gold: -4.98%
Diff: 1644% advantage Bitcoin and Growing
(Here's how I get it, still a WIP to think about whether it differs from dree12's formulation)
def _gcbu(PriceBitcoinToday,PriceGoldToday,N=1):
goldAmt = N/1690.0
goldValue = goldAmt*PriceGoldToday
bitcoinValue = (N/5.4)*PriceBitcoinToday
ratio = bitcoinValue / goldValue
print("the silverbox update (comparison from the beginning of this thread, March 13th, 2012, gold=1690, Bitcoin=5.4):")
print("Bitcoin is %f. Gold is %f" % (PriceBitcoinToday*100,PriceGoldToday*100))
print("Bitcoin: %4.0f%%" % (bitcoinValue-1.0)*100)
print("Gold: %4.0f%%" % (goldValue-1.0)*100)
print("Diff: %4.0f%% advantage Bitcoin and Growing" % ((ratio-1)*100))
return ratio*100
|
|
|
|
another simpler example would be if Bitcoin went from 1 to 1.2.
if i understand your formula, you would say Bitcoin went up 120%. that's clearly wrong as it only went up 20%.
Well, no... I wouldn't say it "went up" without subtracting 100% from my result. That's not really the real issue here. I understand that financial ppl subtract 100% so they can say "went up". You'd do that with my formula too. But actually personally i think that that terminology starts to get really confusing once you pass 100%. I guess the dry way to say what I'm calculating is: "Change in Purchasing Power: Bitcoin vs. Gold: XXXX%". Whether you subtract the 100% or not is getting into english not math. I think dree12's method and mine are the same, except perhaps around the treatment of "up"... |
|
|
|
That wouldn't be likely 90 USD then.I believe someone mentioned that Gox hasn't fixed the glitch on orders in other currencies showing up instead of USD only.
IT was USD Ok Mr Feb 27 2013 registration, this has been discussed lots of times but I'll be kind b/c btc is pushing against the ATH  Mt. Gox ticker converts foreign currency sales into USD and posts them as if they were USD... |
|
|
|
Only the bears seem to be ignoring/not understanding the fundamentals, which is why I guess only the bears have been seeing a bubble since the 20s
I'd say I've seen a couple of small bubbles. The thing is that a bubble deflating in the face of an underlying increase can still lead to a positive gradient. And don't forget, there can be negative bubbles too. Less likely with fedgov running the presses until they glow red hot but nonetheless... The thing is, there certainly are the right conditions for a bubble. Too much cash with nowhere to go. OTOH BTC is undervalued right now so cash may not be inflating a bubble, just inflating a ball to the right pressure. This is Bitcoin in my eyes right now:  nah, more like a slime mold:  No my friends, Bitcoin is the Zerg and this is the Zerg Rush!!! |
|
|
|
...
Sure we understand your method. I haven't spent the 5min to work out whether it is mathematically valid or not because I don't think that's what you really want. I think you really want to ask yourself "How much better off am I since I moved from gold to bitcoin." The script I posted gets you that number because it takes the ratio of the amount of $ you would have today if you had invested in bitcoin vs gold.
...
It would only take 5 minutes, right? One of the things they train us Engineers about is not using more significant digits than are appropriate because it implies a false degree of precision. Thus, it is fairly inappropriate to say 1738% (vs. 1700%) if that is not know to be true to what is actually two decimal places. Not to mention if the formula is bogus in the first place. Any CPAs in the house? edit: add '(vs. 1700%)' for illustrative purposes. WHY are you torturing me with symbolic math!! And WRT sig figs this is not a formal report AND cut-paste is easier then retyping... Key: n = now t = then B = bitcoin G = gold So Cypherdoc's method: (bitcoinsNow/BitcoinThen)*100 - (goldNow/GoldThen)*100 Simplify: (Bn/Bt)*100 - (Gn/Gt)*100 Final: 100* ((Bn/Bt) - (Gn/Gt)) My method: ((N/Bt)*Bn / (N/Gt)*Gn)*100 Simplify: 100* ((Bn/Bt) / (Gn/Gt)) So what do we really want to report? Ok now let's imagine gold is at 1 and hasn't changed at all and bitcoin started at 1 and went to 4. So the bitcoin holder could buy 4 times as much of something as the gold holder. My equation yields 400%, cypherdoc's yields 300%. Which makes sense to you? What if gold falls from 1 to .5 and bitcoin again from 1 to 4. So the bitcoin holder could buy 8 times as much stuff as the gold holder. And my eqn yields 800%, cypherdoc's yields 100*(4 - .5) or 350%. |
|
|
|
the silverbox update (comparison from the beginning of this thread, March 13th, 2012, gold=1690, Bitcoin=5.4):
Bitcoin: +1400%
Gold: -6%
GPL: -36% silverbox long
Diff: +1406% advantage Bitcoin and Growing
This isn't the right way to calculate this. The proper way is to see how many ounces of gold x bitcoins can buy, where x is the number of bitcoins an ounce of gold could have bought. To make the numbers more manageable I will use a gram of gold: March 13, 2012: 1 GAU could buy 54.33 USD or 10.062 XBT. March 26, 2013: 10.062 XBT could buy 809.99 USD or 15.768 GAU. Therefore there was a profit of 1476.8%, far higher than your estimate of 1406%. all i'm doing is comparing their individual appreciation rates against themselves and then each other since 3/13/12: 1. gold: 1690 on 3/13/12 vs. today 1598 or 1598/1690=6% loss. 2. Bitcoin: 5.4 on 3/13/12 vs. today 81 or 81/5.4=1400% gain. Sure we understand your method. I haven't spent the 5min to work out whether it is mathematically valid or not because I don't think that's what you really want. I think you really want to ask yourself "How much better off am I since I moved from gold to bitcoin." The script I posted gets you that number because it takes the ratio of the amount of $ you would have today if you had invested in bitcoin vs gold. And the answer today (using 87.70 and 1598.70) is 1716%By the way, cypherdoc a few days ago there was a discussion as to whether this thread helped any newbies and I would like to tell you that it helped me tremendously. In early 2012 these forums were full of kids, scammers and basically ppl who considered bitcoin to be a game where normal morality and laws don't apply. Your posts gave me hope that the potential of this currency could actually be realized and also affirmed my own belief that gold was fully saturated. To me he is just the opposite of my trader stereotype(that would be S3052), as he is so fanatic and full of love for bitcoin, but when it comes to the price I fully concede defeat, I have never expected it to be near $100 this early, something I only considered possible about a year later. I wasn't referring to his tone but to his content. |
|
|
|
|
Awesome job Trace! I see what you did when you refused to focus on bitcoin's price volatility and instead focused on its utility... brilliant!
|
|
|
|
I had no idea about this stuff. So your basically saying there are services that have human beings breaking captcha's and people pay for them ? lol
Roughly 80% is still done by specialized scripts developed by the de-captcha service operator, then the 20% that script can not break gets sent to humans in Thailand/Philippines or something, and gets manually broken. Can your script (or human) crack my proposal? |
|
|
|
|
+1 to the suggestion to make XBT 1 millionth of a BTC... its unfortunately small right now, but we have the colloquial BTM (millibitcoin) to get us there (and I strongly recommend at 100USD/BTC people start switching over b/c psychologically nobody wants to own a tiny fraction of something)
what feels better? Spend a buck for 10 BTM or .01 BTC
|
|
|
|
|
Necroing this thread because with bitcoins' growth in popularity this may start becoming relevant.
Also, its not just about captchas... news sites are increasingly requiring login even to +1 or flag a comment; this is a serious PITA! I'm guessing that's because of spammers as well.
I wonder if you would really need to send a txn. Imagine a browser plugin that simply signs a message containing the browser's IP address from a bitcoin address that has held X coins for N days (X,N limits can be chosen by each website). This would limit spammers to relatively few accounts and IP addresses. These could rapidly be identified, either because there are a lot of IP addresses using the same bitcoin address (spam botnets) or simply by downvotes.
Of course just like with today's captchas a web service is providing this so end sites don't have to worry about bitcoin. And is gathering spammer data across all sites...
|
|
|
|
Well, after waking up to see the price at $87 I thought that maybe, maybe we might have to slow down a bit. This is an incorrect assumption.
Bids are STILL about 9.5mil USD while asks are below 50k BTC. These indicators have remained unchanged throughout this burst.
The rally is actually incredibly muted, for the reason that there exists an unbelievable flock of sheep-bears, who don't know math, and are willing to sell after every tick up. If they believe it is a great idea to sell <$100, it may take until next week before we see it. Friday is bank holiday in Europe, so the peak for a week should definitely come tomorrow. ^My thoughts exactly. It's actually quite slow-going, particularly compared to last week; with no real walls of resistance in the 80's. Let's put ourselves in the shoes of a newbie. So excited to get some bitcoins! I'll move 4000 to Mt Gox to buy 100 coins!!! Then the big wait... finally the money is there and I can buy WHAT ONLY 45 coins?!?? Sh*t... but look, there was recently flash crash and recovery down to the mid 50s. I guess I'll try to catch that this weekend. tl;dr; its gonna take time for the new money to screw its head around the new prices... |
|
|
|
the silverbox update (comparison from the beginning of this thread, March 13th, 2012, gold=1690, Bitcoin=5.4):
Bitcoin: +1400%
Gold: -6%
GPL: -36% silverbox long
Diff: +1406% advantage Bitcoin and Growing
This isn't the right way to calculate this. The proper way is to see how many ounces of gold x bitcoins can buy, where x is the number of bitcoins an ounce of gold could have bought. To make the numbers more manageable I will use a gram of gold: March 13, 2012: 1 GAU could buy 54.33 USD or 10.062 XBT. March 26, 2013: 10.062 XBT could buy 809.99 USD or 15.768 GAU. Therefore there was a profit of 1476.8%, far higher than your estimate of 1406%. all i'm doing is comparing their individual appreciation rates against themselves and then each other since 3/13/12: 1. gold: 1690 on 3/13/12 vs. today 1598 or 1598/1690=6% loss. 2. Bitcoin: 5.4 on 3/13/12 vs. today 81 or 81/5.4=1400% gain. Sure we understand your method. I haven't spent the 5min to work out whether it is mathematically valid or not because I don't think that's what you really want. I think you really want to ask yourself "How much better off am I since I moved from gold to bitcoin." The script I posted gets you that number because it takes the ratio of the amount of $ you would have today if you had invested in bitcoin vs gold. And the answer today (using 87.70 and 1598.70) is 1716%By the way, cypherdoc a few days ago there was a discussion as to whether this thread helped any newbies and I would like to tell you that it helped me tremendously. In early 2012 these forums were full of kids, scammers and basically ppl who considered bitcoin to be a game where normal morality and laws don't apply. Your posts gave me hope that the potential of this currency could actually be realized and also affirmed my own belief that gold was fully saturated. |
|
|
|
I was making 60%-80% doing arbitrage on eBay. Until the chargebacks came in, luckily I managed to break even after it all. Do.not.bother.
Whats a chargeback? The customer saying you didn't deliver or something and getting their money back? Yeah. This wasn't technically a chargeback (where you make a complaint to your credit card provider/bank and claim a transaction was fraudulent and they reverse the payment). It was a Paypal payment reversal where you claim that your item never arrived and Paypal refund you without question and zero fucks given to their fee paying seller. Ahh, thus revealing one of the best use cases for Bitcoin. If someone sets up a ebay for normal real world items using only Bitcoin that might help make Bitcoin more popular. Actually I wonder how hard it would be to set up a ebay for Bitcoins? EDIT: This exists, http://www.bitmit.net/en/bestsellerBut yes, marketplaces for virtual goods like game items or music have a lot of fraud because a product does not need to be sent to an address and arrived instantly. So if you steal a CC and buy this kind of stuff its very easy to walk away with the goods. If I remember correctly, this was a major reason why ebay stopped allowing these items to be sold. But Bitcoin is the perfect currency for this market. |
|
|
|
the silverbox update (comparison from the beginning of this thread, March 13th, 2012, gold=1690, Bitcoin=5.4):
Bitcoin: +1400%
Gold: -6%
GPL: -36% silverbox long
Diff: +1406% advantage Bitcoin and Growing
This isn't the right way to calculate this. The proper way is to see how many ounces of gold x bitcoins can buy, where x is the number of bitcoins an ounce of gold could have bought. To make the numbers more manageable I will use a gram of gold: March 13, 2012: 1 GAU could buy 54.33 USD or 10.062 XBT. March 26, 2013: 10.062 XBT could buy 809.99 USD or 15.768 GAU. Therefore there was a profit of 1476.8%, far higher than your estimate of 1406%. I think this method is mathematically equivalent to mine posted just above... but can't be sure w/o your formula |
|
|
|
the silverbox update (comparison from the beginning of this thread, March 13th, 2012, gold=1690, Bitcoin=5.4):
Bitcoin: +1400%
Gold: -6%
GPL: -36% silverbox long
Diff: +1406% advantage Bitcoin and Growing
What prices did you use? I didn't spend any time reasoning out your algorithm, but the algorithm below imagines you spent N on bitcoin and gold at the prices stated and then figures out the ratio between the worth of the 2 investments today. Of course, "N" just falls out of the equation so I use "1"... Even if you don't know python, you it should be possible to follow it: def goldCollapsingBitcoinUP(PriceBitcoinToday,PriceGoldToday,N=1): goldAmt = N/1690.0 goldValue = goldAmt*PriceGoldToday bitcoinValue = (N/5.4)*PriceBitcoinToday ratio = bitcoinValue / goldValue return ratio*100 # return percentage cypherdoc, if you actually did jump into linux you can write "python" on the command line, then cut-paste this into the window. Anyway, plugging in 81.75 and 1598.50 yields.... Diff: +1600% |
|
|
|
About the AML/KYC triggers, isn't this only a real problem when your not being honest on your tax filings?
Correct No, not the only issue. You transfer gets held up (in fiat) for weeks waiting paperwork and review. This market could double in that time. |
|
|
|
Have you guys noticed the EUR price at Bitcoin24? It's at 64 right now compared to 60 at MtGox and has been to 69.6 which translates to $89. One could easily make a handsome profit by doing some arbitrage.
How long does it take to withdraw Euros from Bitcoin24 then get the deposit to clear at MtGox? What can happen to the Bitcoin price in that time? For short-term arbitrage, you just need to keep funding the accounts with BTC and $ then flip. You don't have to move the $ around at short notice. Problem is: those EUR markets have been higher than mtgoxEUR for weeks now. When you want to capitalize on this, you should act very fast, not moving euros around. You could for instance send 420 BTC to Bitcoin24 and sell for 63+, gaining EUR 26460. Then simultaneously buy 420 BTC for 61 EUR at MtGox with fiat you have in your account there. You just made 840 euros. A couple of these trades each week and you could give up your job. you need buffers at the exchanges for that. This incurs time value cost of the moneys you have laying around (earning no interest). Assuming it takes 5 days to move the fiat from bitcoin-central sepa account to mtGox account and have it credited. So you had 26,000 laying around for 5 days (not even countain a possible wait for the conditions to be right). So you made 3.2% in 5 days (roughly 0.64% a day). In a bull-market like this, you could've just lent out that fiat on bitfinex and probably gotten more than that. Also don't forget moving such amounts of fiat every 5 days likely triggers some AML/KYC issues and maybe tax problems. Yes this you are exactly right. And remember fiat on an exchange has almost the same risk profile as BTC. Its not insured and an attacker could spend it to BTC and then get it off the exchange. I did some arb and market making across a few exchanges through 2012. It was pretty good during the stability around 5-10. You certainly do keep fiat and BTC in all exchanges, but then you have to rebalance because like we are seeing with the european exchanges these past weeks, a particular exchange will lead or trail consistently over a several month period (exchanges can't be modeled like random walks). And everyone was hitting the transfer limits whenever the markets diverged, which actually was sort of nice b/c it let a lot of players into the game. I did my last arb at 14.xx, and actually missed it -- lost point something % once exchange fees were added in. Now I'm just holding on for dear life!!! :-) (and buying the flash crashes) When you add in the massive reduction in risk (all BTC are in offline wallets) then you can really see why the markets are highly divergent right now. |
|
|
|
|
Unlike tulips, the internet stocks were a bubble but somehow the internet has completely transformed everyone's lives and made a lot of people a lot of money.
Add this bubble chart to a decade+ long term "S" curve (we are still in the very beginning so it is not distinguishable from exponential) and you'll have what I think bitcoin will do.
|
|
|
|
|
Show Posts
