Bid/Ask Spread Tracking

>>2600129
Don't know about stocks, but it would be useful for crypto.

>>2600158
how so? sorry please explain, i'm more of a developer than a markets guy

File: python.jpg (68KB, 640x386px)

68KB, 640x386px

>>2600129
>uses python to script bots
>doesn't realize he's a slave for the moneychangers
>get fleeced all the way down trying to make fractions of a penny over a million times
>realizes his whole life passed him by while some bullshit idea survived for just a few more years
>avoids deep life changing conversations to keep the lie alive

>>2600129
>Is this a worthless endeavor?

Explain a bit more about what you are trying to accomplish.

>>2600667
I'm trying to take a 5-minute grain dataset of the bid/ask spread of each stock in the ASX. I will batch process it at the end of each day to see the average spread.
> continue doing this daily.

If in 2 month's time, the spread goes narrower than average (over an extended period of the day), it's a sign that the stock is moving, either up or down in a big way. this will be my signal of entry

>>2600346
I don't think I'm going to be competing with the HFT's in this. I simply want a way to detect which stock is being heavily bought/sold based on the wideness of the spread compared to average.

>>2600867

OK, now I get it. Here's my advice: if you have been studying the data and this signal makes sense to you and you believe it gives you an advantage, use it. The caveat is that I don't know wtf I'm talking about. I'm a simple mathematician who stumbled into trading as a hobby. How many stocks are there in the ASX?

>>2601189
>OK, now I get it. Here's my advice: if you have been studying the data and this signal makes sense to you and you believe it gives you an advantage, use it. The caveat is that I don't know wtf I'm talking about. I'm a simple mathematician who stumbled into trading as a hobby. How many stocks are there in the ASX?

Gotcha.

>number of stocks:
There's about 2372 symbols that I can track at the moment, so I'm seeing, with 7 hours of trading every day, and at 5 minute intervals, I would be storing about 2372 x 7 x 10 = 166,040 records every day to my sql server db.

>>2601468

Have you prototyped it with a subset of the symbols or are you still in the pondering phase? Sounds like you already know that the pondering phase is important.

>>2601691
I have prototyped it already, like created a script. Planning to schedule it for trials tomorrow in my local. If I can create a time series, I intend to create a CRON job in my amazon EC2 linux box that automates the triggering of said script every 5 minutes.

True - pondering is the hardest part I guess, which is why i'm turning to you guys

OP, this is a pretty good idea, you're essentially looking at volume (this is one thing that influences spreads - higher volume, more "liquid" stock, therefore smaller spread http://www.investopedia.com/ask/answers/06/bidaskspread.asp)

Bearing in mind that you're looking at something which is influenced by volume, perhaps you should also consider having something which looks at ASX volume, I'm not experienced enough to tell you how to do this though.

>>2601730
Wow, what do you know, I didn't actually even consider looking at volume in conjunction with the bid/ask spread.

Ok I'll read a little bit more about it, thanks, anon.

>>2601864

It would be interesting to analyze and check for correlation between volume and the spread. My guess is that it is strong.