Introduction to Zipline: A Trading Library for Python • Hacker News Board

Re: Introduction to Zipline: A Trading Library for Python

#11

Post by howlin » Mon, Jul 25, 2016, 6:44 PM UTC

deutronium wrote: post #4

chollida1 wrote: post #2

I've typed and deleted this post a few times trying to find a way that it doesn't sound kind of pompous but if it helps save one person alot of money then screw it, I'll sound pompous.... I get asked quite a bit on how to start doing algorithmic trading and the first thing I always tell people is don't. I think I've said this many times now but the number of people who come at it with the thinking "I'm a computer sci…

I don't know anything about algorithmic trading, but I'm just wondering, are the people who do make money lots of money out of it, those who have servers close to the data source, who do high frequency trades.

You can make money having by having better access like the HFT firms or by having data not widely available. You can also make money by applying well known principles more intelligently than others. This latter approach usually requires a lot of money. You can't afford retail brokerage costs when you're in a highly crowded and competitive trade.

The best way to make some money as a personal trader is to take advantage of the liquidity premium in one way or another. Because you're trading money in the 5 or 6 figures rather than the 7 or 8 figures, you can take advantage of smaller opportunities without thoroughly distorting the market with your own trades. These smaller trades usually require research and market insight rather than clever algorithms. These trades are usually on financial instruments that don't have a lot of easily accessible data to build an automated trade on top of.

Re: Introduction to Zipline: A Trading Library for Python

#12

Post by dragonwriter » Mon, Jul 25, 2016, 6:45 PM UTC

canttestthis wrote: post #3

chollida1 wrote: post #2

I've typed and deleted this post a few times trying to find a way that it doesn't sound kind of pompous but if it helps save one person alot of money then screw it, I'll sound pompous.... I get asked quite a bit on how to start doing algorithmic trading and the first thing I always tell people is don't. I think I've said this many times now but the number of people who come at it with the thinking "I'm a computer sci…

I never understood this. If its possible to be a profitable independent day trader, and we know it is because many are, then it should be possible to code the rules you follow and become a profitable algo trader.

> If its possible to be a profitable independent day trader, and we know it is because many are, then it should be possible to code the rules you follow and become a profitable algo trader.

There's actually a logical error here: if trading results were essentially random, a certain subset of traders (including day traders) would, at any given time, have profitable records. But you would not be able to derive (and, therefore, implement in code) any set of rules which would make automated trading profitable. You would just either be lucky, or not, as an algorithmic traders just as you would as any other kind of trader.

Now, I'm not saying that trading profits are random, but the existence of some profitable traders does not mean that there are rules you can deduce from their behavior that will guarantee profitable trading when implemented by someone else (either in an automated system or otherwise.)

Re: Introduction to Zipline: A Trading Library for Python

#13

Post by deutronium » Mon, Jul 25, 2016, 6:46 PM UTC

howlin wrote: post #11

deutronium wrote: post #4

Earlier quoted context omitted.

I don't know anything about algorithmic trading, but I'm just wondering, are the people who do make money lots of money out of it, those who have servers close to the data source, who do high frequency trades.

You can make money having by having better access like the HFT firms or by having data not widely available. You can also make money by applying well known principles more intelligently than others. This latter approach usually requires a lot of money. You can't afford retail brokerage costs when you're in a highly crowded and competitive trade. The best way to make some money as a personal trader is to take advantag…

Thanks! That's a very insightful reply

Re: Introduction to Zipline: A Trading Library for Python

#14

Post by lojack » Mon, Jul 25, 2016, 6:46 PM UTC

canttestthis wrote: post #3

chollida1 wrote: post #2

I've typed and deleted this post a few times trying to find a way that it doesn't sound kind of pompous but if it helps save one person alot of money then screw it, I'll sound pompous.... I get asked quite a bit on how to start doing algorithmic trading and the first thing I always tell people is don't. I think I've said this many times now but the number of people who come at it with the thinking "I'm a computer sci…

I never understood this. If its possible to be a profitable independent day trader, and we know it is because many are, then it should be possible to code the rules you follow and become a profitable algo trader.

> If its possible to be a profitable independent day trader, and we know it is because many are

You're forgetting about the many that aren't profitable. You're also forgetting about the majority that are less profitable than the market average. The latter is probably the easiest to overlook. Yes, they are technically profitable, but so are packaged portfolios of stock held for long periods of time. If you can't be more profitable than someone who puts in zero effort, then whats the point of putting in more work?

Re: Introduction to Zipline: A Trading Library for Python

#15

Post by hobo_mark » Mon, Jul 25, 2016, 6:47 PM UTC

chollida1 wrote: post #2

I've typed and deleted this post a few times trying to find a way that it doesn't sound kind of pompous but if it helps save one person alot of money then screw it, I'll sound pompous.... I get asked quite a bit on how to start doing algorithmic trading and the first thing I always tell people is don't. I think I've said this many times now but the number of people who come at it with the thinking "I'm a computer sci…

Can't find the original comment but somebody wrote here time ago that some HFs let you run on their infrastructure (keeping 85/90% of any profits) if you can prove to have a valid strategy. Just out of curiosity, I could not find any other information about this, is it a thing that actually happens?

Re: Introduction to Zipline: A Trading Library for Python

#16

Post by roywiggins » Mon, Jul 25, 2016, 6:48 PM UTC

canttestthis wrote: post #3

chollida1 wrote: post #2

I've typed and deleted this post a few times trying to find a way that it doesn't sound kind of pompous but if it helps save one person alot of money then screw it, I'll sound pompous.... I get asked quite a bit on how to start doing algorithmic trading and the first thing I always tell people is don't. I think I've said this many times now but the number of people who come at it with the thinking "I'm a computer sci…

I never understood this. If its possible to be a profitable independent day trader, and we know it is because many are, then it should be possible to code the rules you follow and become a profitable algo trader.

There are lots of successful software developers too, but writing out a series of rules to automate writing code is not going to happen soon. Unless day trading is substantially easier than programming, an autonomous robotrader that can make money seems unlikely.

Re: Introduction to Zipline: A Trading Library for Python

#17

Post by sjbase » Mon, Jul 25, 2016, 6:52 PM UTC

hobo_mark wrote: post #15

chollida1 wrote: post #2

I've typed and deleted this post a few times trying to find a way that it doesn't sound kind of pompous but if it helps save one person alot of money then screw it, I'll sound pompous.... I get asked quite a bit on how to start doing algorithmic trading and the first thing I always tell people is don't. I think I've said this many times now but the number of people who come at it with the thinking "I'm a computer sci…

Can't find the original comment but somebody wrote here time ago that some HFs let you run on their infrastructure (keeping 85/90% of any profits) if you can prove to have a valid strategy. Just out of curiosity, I could not find any other information about this, is it a thing that actually happens?

Zipline (from OP's article) was created by a company that does something like that: https://www.quantopian.com/faq

Re: Introduction to Zipline: A Trading Library for Python

#18

Post by CaptainKanuk » Mon, Jul 25, 2016, 6:59 PM UTC

chollida1 wrote: post #2

I've typed and deleted this post a few times trying to find a way that it doesn't sound kind of pompous but if it helps save one person alot of money then screw it, I'll sound pompous.... I get asked quite a bit on how to start doing algorithmic trading and the first thing I always tell people is don't. I think I've said this many times now but the number of people who come at it with the thinking "I'm a computer sci…

I run Quantopian's Lectures, which are also intended to teach the statistics behind trading. https://www.quantopian.com/lectures

Re: Introduction to Zipline: A Trading Library for Python

#19

Post by unknown_apostle » Mon, Jul 25, 2016, 7:05 PM UTC

Everybody's trading nowadays.

How about just investing :-)

I.e. focus on periods longer than a year, which so few people/professional market participants do. And on actual businesses instead of the crazy antics of a line.

I wonder if you could use something like Zipline/Quantopian to screen huge amounts of consolidated balance sheets for markers of undervaluation. You could reject 1000s of companies and focus your “manual” vetting on the few that remain.

If you can find the dollar selling for half a dollar and you can understand why it's selling for that price (e.g. because the entire market is down), you may have identified a winner. Then all you need is a little guts and lots of patience. And a predefined set criteria that you would constantly monitor to decide if your thesis is still valid.

Re: Introduction to Zipline: A Trading Library for Python

#20

Post by sjbase » Mon, Jul 25, 2016, 7:05 PM UTC

cloudjacker wrote: post #9

deutronium wrote: post #4

Earlier quoted context omitted.

I don't know anything about algorithmic trading, but I'm just wondering, are the people who do make money lots of money out of it, those who have servers close to the data source, who do high frequency trades.

algorithmic trading doesn't equal high frequency trading. people conflate these a lot even though they know better. high frequency trading requires algorithmic trading. but algorithmic trading can implement warren buffet or suzi orman's style if thats what you wanted.

Agreed, HFT has become a catch-all term for anyone trying to make a marketing/political statement about trading. Like the term "big data" - it's been used to describe so many things that it no longer describes anything.

Another thing that gets ignored is the the difference between a trading strategy and an execution algorithm.