How many of you need to learn coding to work with data ?

[Bad_Badness]

It is not like one has to code everything. There are vast libraries available. Likewise, there are so many examples available too that allow one test to understand and modify to suite a need. These days it is mostly about putting pieces together. Once one sets up the dev environment, it is just a matter of solving your needs.

I think it is misleading to suggest that coding is a do it once and be done situation. Code needs to be maintained, modified, updated, secured, etc. These are ongoing tasks. Suggesting that you are going to have someone do it and be done is a misrepresentation of "coding".

It is more like yard work, get power tools, maintain them, and get it done. Or hire a service for as long as you have a yard.

 

[schizo]

A profitable strategy can be overfitted too. Most strategies are based on mathematics (at least for successful funds) so you need coding. Why do you think that Renaissance/Medallion has a lot of mathematicians and lots of computer power?

That's an anomaly, eg. outlier. You and I will never get even close to Rentec, no matter how much we know about coding.

 

[virtusa]

That's an anomaly, eg. outlier. You and I will never get even close to Rentec, no matter how much we know about coding.

The strategy I use is mathematical and I needed from the first day tools to run own developped formulas in testing. Formulas are code. That requires coding and computerpower.

If you just watch the DOM or draw dozens of lines on charts, you clearly need no coding.

 

[schizo]

The strategy I use is mathematical and I needed from the first day tools to run own developped formulas in testing. Formulas are code. That requires coding and computerpower.

If you just watch the DOM or draw dozens of lines on charts, you clearly need no coding.

So my question to you: Is your mathematical formula any more profitable than a strategy based on dozens of lines on the chart?

I'm not saying coding is unnecessary. But before you even start coding, you need to have a profitable strategy in place.

 

[Good1]

So my question to you: Is your mathematical formula any more profitable than a strategy based on dozens of lines on the chart?

I'm not saying coding is unnecessary. But before you even start coding, you need to have a profitable strategy in place.

Interesting point, because there are really only two major strategies, trend following, and reversion to the mean. You can follow a trend with stops and targets, or you can stay in, always in the market, reversing direction. You can also revert to the mean with stops and targets. This is slightly more complicated with options. For either trend following, or reversion to mean, it should be possible to put lines on a chart, even for statistical arbitrage, which is how i would describe what I am researching. Maybe actual arbitrage and market making is more math centric, but those are technically reversion to the mean centric

 

[virtusa]

So my question to you: Is your mathematical formula any more profitable than a strategy based on dozens of lines on the chart?

I'm not saying coding is unnecessary. But before you even start coding, you need to have a profitable strategy in place.

The advantage of math is that it produces more accurate and consistent results that stay the same no matter how many times you do the math again.
The problem with drawing lines is that it is never as exact or as accurate as math. If I ask you, with an interval of 1 year to draw your lines on a specific chart, you will surely have different lines. They will not be 100% the same when you have to draw them again 1 year later.
Another advantage is that within a wave I have a better view on the probable evolution of the market as I get intermediate points that indicate intermediate tops and bottoms where reversals happen. In some waves (intraday) I have sometimes more than 3 points where I can take profit and get in again at a better price which increases my profit noticeable.

I posted when BTC was around 66K that I had a short signal weekly. At that moment there was not a single line that showed anything. Math is always much faster than S/R lines or other patterns, and gives much less fake signals (which has a huge impact on the returns).

I have seen trades from you in past, and based on their results I can confirm that my results are far better. To be clear I don't just have 1 formula, I have many. They are build in modules and then linked to each other.

If you want to start building a strategy, the first thing you do is calculate, so coding. You need the difference between entry and exit so that is math.

 

[globalarbtrader]

Yes ?
if I give you bulk of data can you pull all stocks that had earnings on the day, gapped 30 %, spiked 10% in the first 30 min of mkt open had a low price in the first 30 min which is lower than the low price in first 60 mins, broke out through the 200 MA, below VWAP, etc etc, all in a few seconds ?

Why would you need to do that in seconds? Unless you're just massively overfitting your trading strategy? Heck, with that filter you're definitely overfitting to hell.

no, research is when you have an entire universe of stock data, and you can query it with words instead of writing code. those are just templates, you can change and improve on top of them by adding filters/more data points.
I don't see how you can "reproduce" any of what i mentioned, since it's a huge DB linked to a complex SQL filtering system.

"Ooohhh. a complex SQL filtering system" give me a break.

To answer your question: I learned to programme way before I started trading (7 years old and 25 years old respectively, or if you prefer 41 years ago and 23 years ago). I currently run a fully automated futures system on 140 instruments using my own code. It's fair to say I'm not your target user.

I think there is probably a place for products that allow you to implement simple moving averages and the like, but you reach limitations on these very quickly, and nearly all the more sophisticated off the shelf products are gilt edged invitations to overfit - like yours.

This rather thinly veiled attempt to push your product is a bit weak and will ultimately undermine your brand. You're a sponsor. If we're interested in your product we'll click on your name. You're much more likely to get value from your sponsorship if you post useful and engaging material, then people will think 'Ah this guy knows his stuff I wonder what his product is like'. Someone like https://www.elitetrader.com/et/members/bone.10535/ would be a good example to look at. He is the ET member with the most followers who is also a sponsor. He has his own thread in the sponsor area he uses for pushing his stuff, but on the rest of the board he's just useful.

GAT

 

[Spikeet]

Why would you need to do that in seconds? Unless you're just massively overfitting your trading strategy? Heck, with that filter you're definitely overfitting to hell.



"Ooohhh. a complex SQL filtering system" give me a break.

To answer your question: I learned to programme way before I started trading (7 years old and 25 years old respectively, or if you prefer 41 years ago and 23 years ago). I currently run a fully automated futures system on 140 instruments using my own code. It's fair to say I'm not your target user.

I think there is probably a place for products that allow you to implement simple moving averages and the like, but you reach limitations on these very quickly, and nearly all the more sophisticated off the shelf products are gilt edged invitations to overfit - like yours.

This rather thinly veiled attempt to push your product is a bit weak and will ultimately undermine your brand. You're a sponsor. If we're interested in your product we'll click on your name. You're much more likely to get value from your sponsorship if you post useful and engaging material, then people will think 'Ah this guy knows his stuff I wonder what his product is like'. Someone like https://www.elitetrader.com/et/members/bone.10535/ would be a good example to look at. He is the ET member with the most followers who is also a sponsor. He has his own thread in the sponsor area he uses for pushing his stuff, but on the rest of the board he's just useful.

GAT

I'm not trying to push anything, but rather understand the community's need and what traders think about programming. That's why i started this discussion.

in your case - you started to code way before, but what if you started today and didn't know coding ? and had a strategy in mind to implement ? wouldn't you want a quick way to implement it, assuming it can handle your strategy complexity

 

[globalarbtrader]

I'm not trying to push anything, but rather understand the community's need and what traders think about programming. That's why i started this discussion.

in your case - you started to code way before, but what if you started today and didn't know coding ? and had a strategy in mind to implement ? wouldn't you want a quick way to implement it, assuming it can handle your strategy complexity

I suppose, but there are already dozens more platforms that can already do that.

GAT

 

[Spikeet]

I suppose, but there are already dozens more platforms that can already do that.

GAT

platforms that are no-code and where you can source data, build strategies and run them ? can you share some for the community ?