FiftyEight Fluffs Around In Futures

[fiftyeight]

The price keeps going up.

For continuous contract and historical data esig wants another $14 US / month. $161 US/ month total. This seems the only way to trade the Nikkei on the OSE with NT.

An expensive price to pay to trade a convenient time zone when I am only home half the month......

 

[ThingyMajiggy]

The price keeps going up.

For continuous contract and historical data esig wants another $14 US / month. $161 US/ month total. This seems the only way to trade the Nikkei on the OSE with NT.

An expensive price to pay to trade a convenient time zone when I am only home half the month......

Try CFDs? No fees and no comms, just the spread. Might be an easier/cheaper option.

 

[CanOz]

The price keeps going up.

For continuous contract and historical data esig wants another $14 US / month. $161 US/ month total. This seems the only way to trade the Nikkei on the OSE with NT.

An expensive price to pay to trade a convenient time zone when I am only home half the month......

Does that include level II?

 

[fiftyeight]

Try CFDs? No fees and no comms, just the spread. Might be an easier/cheaper option.

I can execute through AMP easier enough, its the data to chart VP stuff causing the issues. Been trawling a few other forums and most ppl seem to have the same issue.

Think I will bite the bullet and just switch to Mulitcharts. AMP have a free version so will give it a crack

 

[fiftyeight]

Does that include level II?

Yeah. I think haha I have been talking to so many ppl.

Nightshift makes talking to the US much easier

 

[fiftyeight]

Best I ramble in my own thread.

For better or worse I am going with Python.

A few reasons, for a beginner there are plenty of good resources so even though the learning curve maybe a bit steeper than AB it shouldn't be an issue (famous last words). The last time I did anything like this was 10+ years ago at uni and I paid little attention so I am starting from scratch, I could do with going over some computer basics. I also need to go over some of the basics again anyway of how data is structured, stored and accessed. Same with my math and stats, it has been a long time since I have used any of it so having to look at the code a bit deeper will be a good refresher.

I would like have a look at some betfair data and how that applies to the bigger bookies. More for interest than expecting anything to come of it.

Not in a pessimistic way but I not super confident in "cracking it". There are a lot of people who are lot smarter, have more time, more training and more money than me looking at this....both in trading and programming. So if nothing comes of trading the skills I learn with Python will be transferable if I decide to go back to uni a pursue something else.

In saying I am not confident I have been skyping with someone quite involved in the quant side of things who has inspired me to look in to it. Many screen shots of equity curves that look fantastic. I have no secret sauce, but they really dont seem to be doing anything original. All the ideas they shared I have come across before, maybe its all in the application as tech is always saying. I am skeptical as a screen shot of a pretty equity curve mean nothing but everything I will look at initially will help with my intra day (when I actually have time)

Enough time wasted deciding, lets get on with it

 

[fiftyeight]

A few different people have recommended "Learning Python The Hard Way" so I am beginning with that.

He uses Atom but I have already been playing around with the Anaconda install and using Jupyter Notebooks so Ill stick with that.

Was having a look last night at Quantopian.

https://www.quantopian.com/posts/trading-strategy-ideas-thread

Links to many different papers

 

[History Repeats]

sentdex on youtube is very good. Where i learn from.

 

[Trembling Hand]

sentdex on youtube is very good. Where i learn from.

Not if you want to do Time Series Analysis correctly. His approach is fundamentally flawed. In fact it's not flawed it just completely ignores it!!

 

[fiftyeight]

First bang head against the wall moment....the pleasures of learning at home

 

[fiftyeight]

This will teach me for not following the book exactly.

Firstly Python installed wrong and I could call it in Powershell. That is now sorted.

Now I am working in Jupyter and saving in .ipynb which was not an issue....initially. But now I need to execute in Powershell (as per the video) but I cant figure out how to run it.

Not sure if I need to convert the file or call another program instead of python

Seem clunky if I am always converting stuff. The videos I am following use "Atom" and they are saving in .py.

I assume I have missed something really basic here but

Any help would be sweeeeeeeeeeeet

 

[fiftyeight]

That makes me want to cry.

Found the fix, a very simple one 1 min after posting the above. There goes 2 hours of my life I wont get back haha

 

[History Repeats]

Not if you want to do Time Series Analysis correctly. His approach is fundamentally flawed. In fact it's not flawed it just completely ignores it!!

just for the intro and using stuff like panda,ect. I think its a good place to start. But you're right.

 

[fiftyeight]

I have been following the Pandas sentdex videos when I want a break from LPTHW.

He smashes through things, I can just keep up then struggle when I try apply it with out having to refer back 1000 times haha I understand the code, but I just cant recall it that well yet

Having a day where I am questioning my choice, have made minimal progress and can still do more in Excel haha

 

[ThingyMajiggy]

He smashes through things, I can just keep up then struggle when I try apply it with out having to refer back 1000 times haha I understand the code, but I just cant recall it that well yet

That's basically what programming is, looking up documentation, syntax and Stack Overflow

 

[fiftyeight]

Haha yup.

Just simple stuff like working out what the gap between close and open the next day is time consuming.....still figuring it out

 

[fiftyeight]

Gave up on an elegant solution and just shifted the column down 1 row haha

 

[Trembling Hand]

df.open - df.Close.shift(1)

 

[fiftyeight]

df.open - df.Close.shift(1)

Much prettier than mine

df["s1"] = df.Settle.shift(+1)

df["Gap"] = df["Open"] - df["s1"]

 

[Trembling Hand]

Much prettier than mine

Here you go,

https://becominghuman.ai/cheat-shee...learning-big-data-678c51b4b463?imm_mid=0f57e6

Collections of cheat sheets for everything from Python and Pandas basics all the way to Neural Networks.

Get on board