»Fast Profits Daily by Equitymaster

On This Day - 16 NOVEMBER 2021
Using AI in Your Trading System

Brijesh Bhatia, Research analyst

Artificial Intelligence or AI has revolutionised the world of trading. Every serious trader is either thinking of learning it or has already taken the plunge.

But is it helpful in making you a better trader? Is it really necessary to learn AI tools to make more trading profits?

Find out in this video...

Let me know what you think in the comments. I love to hear from you.

Hello viewers. Welcome to the Fast Profits Daily. Myself, Brijesh Bhatia.

This video is focusing on Artificial Intelligence or AI what we say into the technical analysis. Well, AI has taken over the world into new dimensions. Everyone in the world is now having mobiles, smart mobiles or small devices are keeping the manual work on a lower scale for the human beings.

Well into the technical analysis, and there are areas, points that one should understand into the technical analysis for AI. One is the coding, which is the most important part into the AI. Python being the language where most of the people, go into the technical analysis. Julia is the new one which is the advanced level to the Python.


As a technical analyst or as a trader, should you learn this kind of coding system Python or Julia and look for trading on technical analysis? Well, that's always a debateable topic. Some firms hire the coders to code their systems for them. But when it comes to the retail traders, do you really need this coding system? Should you learn these coding languages like Python?

I remember that when 20 years or two decades ago, my brother was studying into the Microsoft languages C++ or the Java, which is now taken over with the more advanced languages. So sure you learn this and what kind of knowledge you do need?

Well into the YouTube world or the internet world all kind of videos are freely available for you to learn this coding, but it's not easy task to code each and every stuff you want. Now, if you look at, apart from Python, if you just go to the Google and type AutoML, you will get a huge library from Goggle coding which will really help you to generate your own trading system but still for me or any kind of layman, that system or that coding language could be very much difficult coming into the writing on that language.

So I believe that see, that is always a debateable should you learn to code your system or should you hire? If you are able to afford hiring, there are a number of freelancers available in the world, who can help you code your studies or your systems.

Well, we have software now which can identify the patterns into the systems. The patterns like head and shoulders in technical analysis, the patterns like harmonics, patterns like even if you look at the Elliot Waves, which if you will just code it, then it will automatically show you on your screen that xyz pattern is forming, or triangle is forming, or a channel is forming, auto trendline system is available. But do they really help you?

I believe at someone of time, yes. It helps you out to identify some kind of pattern which is developing but again if you manually plot the the head and shoulder and the auto systems which plots your head and shoulders, then there won't be much difference. But when you are plotting it manually, you understand that probably, the system has worked in X, Y, or Z stock.

For example, if you can't be Nifty, if you go the back tested results, head and shoulder on the Nifty has very low success percentage. If I look at the triangle structure on the Nifty or the channel structure on the Nifty, even short term and the long term, it has a higher success percentage.

So you know, such kind of systems when you do manual testing or even you when you do the automated back testing, it really helps out. So you need to understand what kind of system you're developing.

There is always a myth that technical analysis is a tool to trade. Again, as I said basically if you look at the technical analysis, it's just one piece on price. Everything is calculated based on price. So if you don't believe in technical analysis, which means you don't believe in the price

As I said, historically, if you look at what is technical analysis. It's the data, the price action data, which is used to predict the future prices. So if you don't believe in technical analysis, it means that you don't believe in the price of the stock, the price of the asset class or it x, y, z. So this plays a key important role.

As I said, if you look at the difference between Algos and Artificial Intelligence into the technical analysis, then technical analysis is really helps you to identify patterns. Algos are basically used for execution purposes. Now, if you know the technical analysis, if you know the systems which you are developing, then it will definitely help you to execute your systems.

What are the advantages if you're looking at artificial intelligence or machine learning or the algos?

The most important is you don't need to scan 1,000s of stocks manually. The system will throw the stocks at you. It will really help you out. Do it. Second is the coding as a ranking system. If you look at the back testing numbers, as I said, head and shoulders has a slightly less percentage. So you can rank head and shoulders on the Nifty at the lower level. You can rank channel or the triangle at the higher level on the Nifty. It will really help you.

So ranking systems here helps a lot, when you want to do I would say the machine learning or AI. So it will throw you that the Nifty pattern is at x, y, z rank and if that systems help you to rank from the higher side, it's definitely a good trading set up, which you have ranked as per your experience or as per the back tested numbers.

Again, the AI definitely helps you to do the back testing. There are a number of people I know who are back testing in excel and they know how to use excel to the macro levels as well. So you need to understand that as well.

Multiple analysis can be done using the machine learning because that can't be done manually. For example, you want to find outperformance Tata Motors against the Maruti, so it will definitely look or show the structure when Tata Motors has outperformed, when it has underperformed against the Maruti. When Maruti has performed against the auto sector. When Maruti performs against the Nifty. So on a click, you will definitely find it easy to access this.

So should you look for or learn this coding language I would say it is not necessary because most of the software available now helps run the easy technical query. There are some of the software, which would like to name it out. Depending upon their pricing, you can choose and just have a trial.

Bloomberg is the one which is on a slightly expensive side.

Amibroker really helps you out to implement your studies into the software. Again you have to buy this software and you have to buy the data for Amibroker. What kind of markets whatever kind of markets you are trading into. Plus, you have a number of libraries available for Amibroker systems.

There are a number of people who are doing freelancing on coding on Amibroker in case you want it at a slightly lower price, depending on the area. So I have done a couple of queries. I've done for myself charging somewhere around Rs 2,500-3,000 and it really helps you out. So you can use Amibroker to code your studies.

The third is IRIS+ from Spyder which is costing some and around Rs 55,000-60,000 and that could be again, it has a technical query in-built which you can add up and so you don't need that coding system for you to throw the stocks in front of you in the scanning.

Tradepoint is the one software, which is available, I guess around Rs 30,000-35,000 per year which helps you to find scanner and they most advantageous point of Tradepoint is that it helps you to back test data as well. So if you're running a system, you can do the back testing of that system how it has thrown a number of trades, the profits, the loss, the strike rate, depending on it you can tweak the systems you like.

So there are some of the softwares which will really help you. Trading View is one which is again where you can code, your own system or on your trading parameters, and it will show you the signals of buys and sells.

So these kind of softwares are available where you don't need to learn Python, you don't need to learn Julia, and you don't need to learn macros of excel. They are easy to scan a number of stocks for you. So again, this is not promoting any software for you.

If you want to learn machine learning or machine trading, there are a number of algos. We have done a video on algos indicating that a number of service providers are there that can help you to create your own algo and execute on that systems. So don't need to implement cost heavy technologies into your trading arena for coding or algo systems.

So signing off. Brijesh Bhatia.

Warm regards,

Brijesh Bhatia
Brijesh Bhatia
Research Analyst, Fast Profit Report
Equitymaster Agora Research Private Limited (Research Analyst)

Copyright © Equitymaster Agora Research Private Limited. All rights reserved.

Any act of copying, reproducing or distributing this newsletter whether wholly or in part, for any purpose without the permission of Equitymaster is strictly prohibited and shall be deemed to be copyright infringement

Disclosure & Disclaimer: Equitymaster Agora Research Private Limited (Research Analyst) bearing Registration No. INH000000537 (hereinafter referred as 'Equitymaster') is an independent equity research Company. The Author does not hold any shares in the company/ies discussed in this document. Equitymaster may hold shares in the company/ies discussed in this document under any of its other services.

This document is confidential and is supplied to you for information purposes only. It should not (directly or indirectly) be reproduced, further distributed to any person or published, in whole or in part, for any purpose whatsoever, without the consent of Equitymaster.

This document is not directed to, or intended for display, downloading, printing, reproducing or for distribution to or use by, any person or entity, who is a citizen or resident or located in any locality, state, country or other jurisdiction, where such distribution, publication, reproduction, availability or use would be contrary to law or regulation or what would subject Equitymaster or its affiliates to any registration or licensing requirement within such jurisdiction. If this document is sent or has reached any individual in such country, especially, USA, the same may be ignored.

This document does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of individual subscribers. Our research recommendations are general in nature and available electronically to all kind of subscribers irrespective of subscribers' investment objectives and financial situation/risk profile. Before acting on any recommendation in this document, subscribers should consider whether it is suitable for their particular circumstances and, if necessary, seek professional advice. The price and value of the securities referred to in this material and the income from them may go down as well as up, and subscribers may realize losses on any investments. Past performance is not a guide for future performance, future returns are not guaranteed and a loss of original capital may occur. Information herein is believed to be reliable but Equitymaster and its affiliates do not warrant its completeness or accuracy. The views/opinions expressed are our current opinions as of the date appearing in the material and may be subject to change from time to time without notice. This document should not be construed as an offer to sell or solicitation of an offer to buy any security or asset in any jurisdiction. Equitymaster and its affiliates, its directors, analyst and employees will not be responsible for any loss or liability incurred to any person as a consequence of his or any other person on his behalf taking any decisions based on this document.

As a condition to accessing Equitymaster content and website, you agree to our Terms and Conditions of Use, available here. The performance data quoted represents past performance and does not guarantee future results.

SEBI (Research Analysts) Regulations 2014, Registration No. INH000000537.

Equitymaster Agora Research Private Limited (Research Analyst) 103, Regent Chambers, Above Status Restaurant, Nariman Point, Mumbai - 400 021. India.
Telephone: +91-22-6143 4055. Fax: +91-22-2202 8550. Email: info@equitymaster.com. Website: www.equitymaster.com. CIN:U74999MH2007PTC175407