October 23, 2019 100

Stock Price Prediction | AI in Finance

Stock Price Prediction | AI in Finance

100 Replies to “Stock Price Prediction | AI in Finance”

  • Amir Rocker says:

    Thanks a lot Siraj. You are the best! keep going please !

  • InterAct Party says:

    speed of the slide over a period of time to comparisons

  • shahbaz ahmed says:

    love from india brother. Keep coming up with great content.

  • karimuzzaman sunny says:

    Just yesterday i saw your video, i think you serve us and its great.. if our teacher can teach like you that could be better ,but its not actually happend.

  • Jose Saldivar says:

    Thanks Siraj. please do us a video on your résumé or better a true C V. Curriculum Vitae.
    Please, . Jose

  • Ian Finn says:

    Hey Siraj, everytime i go to try to implament and use your code for myself, i constantly run into bugs and errors. Could you possiblly tell me why? Additionally i was wondering which interprater you use (the one that says scary bugs mac)

  • Saurabh says:

    This is the best video I have ever come across in AI. I tried many a times to learn ML and AI but almost everytime i left it in between because i was unaware the possibility and application of it, what can i do after learning it.. and here in your video without diverting a single sec, i can trace my roadmap for next 10 year. Thanks of it

  • Brandon Ha says:

    Oh… i want to make it on my mac. How can i start it.. btw i love his videos

  • Marciel Fonseca says:

    How would it do with binary options? It doesn't need to predict the price but rather if it is going up or down for the next x time.

  • MODESTO says:

    I think what largely effects stock price is supply and demand as with anything for sale in the business world. So how can anything human or AI predict how many people around the globe are going to buy a specific stock at any given moment?

  • Ross Frankenfield says:

    I cannot get the app.py to run since some modules were depreciated and I cannot figure it out to save my life. haallp

  • Zes says:

    nst as everything

  • Frank Dobbs says:

    We dont want to predict it.. we want to make money from it.. from real markets.. not simulated.. this is a complete fail, make it more applicable to real life

  • Manish Joshi says:

    I have done CFA course ,apart from that I want to know there is any good course AI in finance or quantative trading .

  • rashed saeed says:

    How do you get A.I. For trading ?

  • SalRite says:

    What are the areas of Mathematics required to master Reinforcement Learning?

  • Cyberer B. says:

    don’t mislead people. The AI has failed to predict stock price. And 7.7 billion investment in AI means you’ll hardly get any job. Yet it does well in trend prediction which is not enough.

  • Vala Zeinali says:

    I did all of this and I got the error: "DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
    "This module will be removed in 0.20.", DeprecationWarning)"

  • Pablo Aguirre says:

    Excellent material, thanks

  • Jewel afraim says:

    How much money u make from stock market using all these stuff

  • PT Batmax says:

    ai is doomed…

  • Mohammad Derakhshi says:

    Hi siraj
    Thanks for your time. I have a question: suppose that we have predicted the stock price. Next, some prominent bodies ( such the President of USA talking about the related issue that could affect the price of that stock wherein has not considered before) how we could consider this new information to the prediction?

  • Nicholas Gans says:

    AI in Crypto!

  • ty10r says:

    How can I learn more about AI uses for a CFO that you mention?

  • Ahmed Albayati says:

    Eat Dicks

  • Skeptis Trader says:

    I am in futures trading and would love to meet a professional machine learning programmer who is interested in doing work and potentially being a business partner

  • 369 says:

    If you really know how to createyour own a.i it’s scary what you could do in life. You are literally a God

  • Matt M says:

    If anyone is good at AI and has a trading solution built with signals holler at me [email protected] I have a good offer for you if your AI is legit 🙂

  • Arun sha says:

    can someone tell me how to add sentiment analysis result (which is good or bad) to row for regression

  • TEE LEE says:

    This is an excellent video. Whether we can download your handouts for this video just for learning?

  • 강성호 says:

    No special metion on Prediction of stock price with AI… disappointed

  • Jignesh Jetani says:

    Mr. Raval are you from gujarat because I like such AI can only think by Gujarati ,it very nice explain

  • Maxi Tube says:

    nodejs needed for this script?

  • Debi Rath says:

    Hi Siraj. I am not able to run the code some how.

  • rinesh mukherjee says:

    Really impressive analysis. Could you please provide the ppt or pdf from where you are showing

  • Streamingmadman says:

    Does the AI's overestimation of the data result in many errors, particularly when it comes to fraud detection?

  • văn lý trần says:

    Hi, every one,
    Every one here who have already did the " Prediction car speed" can have me to do it, or give me some idea to do it. Thanks so much!!!

  • Cesar Augusto says:

    I don't thin LSTM's are great for time series…

  • Ali Pardhan says:

    IMHO the video began at 10-13min mark.

    im here for the time series Prediction itself, not usecases.


    I subscribed your channel some months ago, I wanna to learn AI.. please guide me how can I develop a stock prediction sw for Indian stock market

  • 98thpixel says:

    Nice video. I've seen that one of the most common mistakes people do when attempting this type of predictions is the errors related to "Look-ahead bias". What are your suggestions to minimize the possibility of errors of this type?

  • Zach Kim says:

    16:50 can we really sentiment analysis results transform into numerical variable? seems to me not all 1/0 has to be accounted for same degree of "good" or "bad"

  • Bruce Levy says:

    Algorithmic trading has nothing to do with 10k 8K or whatever sentiment analysis reports you want to scrape up from the internet. Everything is based on market liquidity for trade facilitation. The smartest AI will use this sentiment analysis to see where other AI bits accumulated positions. Then squeeze them out.

  • Conner Swigart says:

    The law of standard distribution is all you need to know about trading. if you understand that you'll understand why day traders loose most of the time. XD

  • Bronze says:

    Dude, genius. Subscribed!

  • A Bazdar says:

    … Great !!

  • Dr. Phil says:

    I both believe in the weak,-semi and strong version of the market hypothesis. The weak version I believe is rare, but does happen. The semi-strong version is less rare, but i belive it usually only (sometimes) applies to smaller, less traded, companies.

    There are active traders, and there are passive traders. I belive in what both do, the passive trader will create diversified portiofolios with a good return-risk ratio. For this task only simple mathematics is needed, and I cant see any need for AI. However data collection algorithms can help find the best portifolios.

    So, active trading is the field of the AI. Here is what you gotta realize: if the semi-strong market hypothesis is true (which it will be in most cases) the data the AI gathers won't be relevant, the market will already have acted on this information, and the price will reflect this. And wheter you make a profit or not will be down to luck.

    Good active traders gathers data that are not publicely available. A short seller might for example look for fraudulent practices by analysing their financial statements etc. and even doing real life spying on the company to gather information. Bet on the stock going down, then release the information. Other active traders will buy enough shares so they can affect how the company is run, if they see a problem (and the solution) and it is the leadership – buy low and sell high.

    Now, how the heck is your machine learning algorithm going to compete with this? it can't. How do the big boys make money on it? through influence, having the right buy code on the stock exchange, and the fastest network connection. Well, unless happen to stumble upon some fact about the inner workings of the market that is not yet known, which is highly unlikely. And its not likely that an AI is going to find such a thing either, since the stock market is a complex system whith trillions of variables that change and all affect eachother.

  • Elvis Hlongwane says:

    Hi Siraj, Good work. I have learnt a lot from you since I have started following your material. You referred to the web template that we can use, where can I download the source code? Thanks in advance.

  • panchi shah says:

    I am not a programming wiz person. But I have UG in financial markets and mba in finance. I have always been interested in so ai also trading and ML. I created my own eth miner. In order to repurpose I thought of this. May be this is a waste of my time. But maybe some career advice is needed. Thanks. Really appreciate it.

  • pravin18in says:

    Absolute crap. To solve a problem, first try to understand it.

  • Vision says:

    Hi…i like your videos.
    Would you suggest me best algorithm for intelligent advising.

  • João da Mota says:

    Hi Siraj, how do I get this WebApp to take a look? I would like to get more information about your work…. Do you have an online course of some kind? Many Thanks!


    Very good presentation, I am entrepreneur in the Insurance Industry with a good knowledge of Management. Anyone with a tech skills would like to work with me ?

  • Misthema says:

    Blablablabla inhale blablablabla inhale blabla inhale blablablabla inhale blabla inhale blablablabla inhale blabla inhale blablablabla *inhale*…. fuck that's all i hear now. What about acoustic panels? 😀

  • паша зайцев says:

    You are sooo smart… And so stupid) You can't predict it. And never will.

  • Shreyas Kulkarni says:

    Why does'nt everyone use a stock prediction AI and become really rich?

  • Ahmed Houza says:

    Wawou you are very smart and handsome man. Bravo Siraj

  • น้องคริปโต CryptoOnline says:

    Fx Ai Robot Arbitrage Trading
    Artificial intelligence Investment management
    👍 No need to watch the screen. Withdraw money every day
    🔜 would like to tell the impression
    Ai with Ai Ajriya Robot
    💰Arbitrage trading💰
    Make 1-2.5% profit per day
    📚 Easy, easy, no need
    😆 Watch the screen for eye strain !!
    Interested in clicking the link to study the system
    🔜 Free express before the deadline


    📌 Click on this link 📌


    [Ai Fxtrading Crypto]] You are invited to join LINE Square.
    Blockchain & Ai Channel to receive information on Lyad >>>

  • shubh Tripathi says:

    is there someone or any stock market company using AI .? if u know so pls tell me brother

  • Texas Mike says:

    good stuff! love it!

  • Mike Sales says:

    is this person having a mouth diarreah?

  • dave stawak says:

    y dont u actually make the ai

  • Vladimir Knyazev says:

    Очень нервный индус. Рассказывает так, как будто себя хочет убедить, что он бы сам у себя … взял-бы, да бы мешает.

  • Kirill Bezzubkine says:

    Didn't like the vid. It's about everything and nothing at the same time

  • Subham Sarkar says:

    15:23 The Satisfying Key Stroking Sound

  • Mega Players says:

    i can hear him breath heeh every 5 seconds

  • Jered Willoughby says:

    I watch your videos repeatedly to learn from your AI progam code. Thanks a million times!

  • Code Cleaner says:

    Pay attention to some basic concepts when trying to predict stock prices:


  • Jab Hutt says:

    There's some kind of mass hysteria of stock markets over a year now, it's left and right, everyone sells/buys/develops. Or is it just google manipulating me into thinking it?:) Or the Architect himself?:))
    Anyways, ran the code and it works! Bought Siraj now with his channel, taking down the videos in the next week! Enjoy them while you can!:))

  • BOB YETUCOM says:

    Great Job.

  • sask1930 says:

    Re Sentient Technologies Hedge Fund – https://www.bloomberg.com/news/articles/2018-09-07/ai-hedge-fund-sentient-is-said-to-shut-after-less-than-two-years

  • Maisonier says:

    If one wants to start studying all this … we need to know all the basics on programming, different languages, Object-oriented programming
    , etc? or what are the basics to start studying this?

  • Blah Bla says:

    Your videos are awesome, it would be nice if you could do something about voice recording audio has noise and breath , you could decrease sensitivity of mic or probably use AI to clear the noise from audio recording

  • Arnoud Verstraaten says:

    Highly interesting. Thank you so much for sharing, Siraj.

  • Japneet Singh says:

    Installing from " pip install -r requirements.txt " throughout : " ERROR: Ignoring numpy: markers 'python_version == "3.5"' don't match your environment

    Ignoring numpy: markers 'python_version == "3.6"' don't match your environment

    Collecting wheel "

    Please help

  • Nico Roets says:

    Total waist of 24 min.

  • OuYang Hao says:

    can anyone tell me where next videos of this series are ?Thanks a lot

  • D B says:

    an idiot teaches even bigger idiots to 'JUST PUT IN ONE LINE AND NOTBOUT ALL THESE MAGIC NUMBERS'. Pathetic

  • Robert Ferraro says:

    Sorry buddy – you cannot, I repeat, cannot predict stock prices. Stock market charts are purely random, with a very slight bias that is exactly that of the market indexes. Even the smartest AI cannot predict a stock price and if it can you will find that it lags, just like the best technical indicators do. So in the end it is no better than the best indicators and a human taking a good educated guess. Go and get yourself some historical data and manipulate it any which way you want. Most (90%) of the time behaviors and price movements are random. No better than flipping a slightly biased coin. I created my own simulation of buying and selling stocks 15 years ago using the proven best indicators and strategies. 10,000 trades… the brokerage house made $100,000. I made $4,000. The only way to truly make money in the stock market is BUY and HOLD.

  • horse drum says:

    They are few men who are expert in stock market trading i count Mr George jorkins as one because of his amazing strategy

  • Flavio Lopes says:

    Very amazing video!! Thanks a lot! I was know the gekko and trading strategies, and this is wonderful strategy. I would like to make a suggestion to you add an another video to playlist Reinforcement Learning for Stock Prediction. Thank you one more time @Siraj Raval. I'm working with Head of Sales Engineer at Semantix an Big Data and AI company, and I'm a contractor too. I would like to talk with you about oportunities. Please send me your linkedin to more easly contact?

  • Ruben Perez says:

    no way I could stand 24 minutes of this fag, couldn't even do a minute

  • Mike A. says:

    If seraj knows how to predict shit he would be 24/7 setting down predicting and not upload stuff like “ how he does it “ lol + I don’t trust ppl with colored hair especially men 😀

  • Sal Calvanico says:

    Talks a lot, 15 minutes, get to your point

  • NitroThe X says:

    Rm -rf*

  • Gustavo Martinez says:

    A video that explains the current methods to understand why reinforcement learning or deep learning has taken that decision, ( causality) would be great!!!!. You are a very inspiring person, maybe we have a lot of differences, but you inspire a lot of people, so that's great!!!

  • Himanshu Sharma says:

    Siraj, please share the link from where you were explaining, it would be really helpful.

  • Elia Giacobazzi says:

    Great video!

  • Farzher says:

    hello comments. i need some help.
    i'm trying to make a trading bot that takes live market orders as input data, and simply predicts when to buy/sell based on that.
    it should learn to buy into breakout trades, reversals, etc.
    what's the standard way/model to do this?

    new to machine learning. the tutorials seem to be mostly about labeling input images.

  • Fuggles says:

    "I'm releasing this because I'm making enough money…" no you are releasing this because it is primatave in comparison to those used in hedge funds. This is "teach people the basics" grade, rather than "implement billion dollar trading fund" grade, and since teaching people online, not managing a billion dollar hedge fund, is what you do, that is why you are sharing it. ; )

  • Heinrich Peter Maria Radojewski Schäfer Leverkusen says:

    He performed a Tesla two-handed frantic waving.

  • Gerardo Moscatelli says:

    Linear regression of a price as a function o time, everything goes up forever until it doesn't 🙂

  • Ean Eisenzimmer says:

    good content! thank you.

  • Ar Gon says:

    Good video. I am putting together a tool to get the senitment of the news / press releases for each stock in my watchlist. Have it run hourly and alert me for positivity. The goal for me is to be able to hone in and buy on good news.

  • Devashish Soni says:


    can you please share the link form where you got stock price data of companies

  • Aleks Martini says:

    This was a pile of horse shit

  • 최소소 says:

    Can you predict international oil prices in a linear cycle?

    Which program should I enter?

  • Sonny Forelli says:

    Wow A Indian speaking a english that I can understand.
    only by that, you deserve a like!

  • kotov dot in says:

    Already outdated with bitcoin killing the game

  • imran khan says:

    When i am doing forecasting for timeseries data the output that i am getting seems to be one-time step lagged with actual output i.e. what we called persistent model. I have tried several ML as well as DL algos as well as statistical algos. What to do in this case. How to correct it??

Leave a Reply

Your email address will not be published. Required fields are marked *

Recent Posts
Recent Comments
© Copyright 2019. Tehai. All rights reserved. .