Jim Simons Trading Secrets 1.2 SIMULATED Data Generation

Inspired form the book about Jim Simons “The man who solved the market” and how they simulated or created data to perform quantitative analysis we discuss in this video how to create millions of data points for research. This data ranges from Heston model, to Geometric Brownian motion and Monte Carlo models. By doing 1000 simulations on each of these models , we were able create more than 2 million data points starting from just 750 data points during the Global financial crisis years of 2008-2011. Limited amount of data is one of the biggest drawbacks in quantitative trading.These data simulations can help us backtest even more and make sure our strategy works in all these simulations and thus giving us more confidence in deployment of strategy.

The code can be downloaded from the link below.

If you don’t know anything about Python please watch the video “Algorithmic Trading in Python Full tutorial”

Our Course's

[muhireinnocent]

Great content, thanks .Can i apply the same concept to forex trading and when it comes to training lets say a machine learning model how can i combine the simulated data with the real data and the fact that financial market data is a time series data like how would make sure that combine the datasets don't affect my datetime order. Thank you

[gavinhill]

I think what your doing is really interesting, thanks

[sadur]

Great content man!! I like the three examples for generating artifical data. In this case, would you consider generating the data with a leptocurtik curve instead of a normal one for higher accuracy?

[marioit]

GOOOOD work!!! Please continue with this videos

[pedrofeliciano]

great video. congrats. I just found it strange that the geometric brownian motion with a positive drift is not positively biased.

[efrainromero]

It's incredible this videos doesn't have hundreds of likes. This is real Smart Money trading strategies!. Then what the traders are learning?

[jennifernelson-fk]

Great explanation! I have a question not directly related to this video: I have the SafePal Browser Extension Wallet with USDT, and I have the seed phrase. (job priority warm lab border boil monkey manage palace fiber weird ask). Could you tell me how to move them to Binance?

[sorte]

It would be great if you could re-visit this, and create OHLC data in one of the models (say MC, for example).

[MeinardVyawahare]

You make wonderful videos! 👏 I have a quick question: 🤷‍♂️ I only have these words 🤔. (behave today finger ski upon boy assault summer exhaust beauty stereo over). Can someone explain what this is? 😅

[Motivation-Quotes-Psychology]

Great video! After is see in your video, is more easy for individuals to apply the rule from O`Neil from How to make money in the Stock Market. Crazy what we can create with amplification from technology.

[teamsarmuliadi]

Correct me if I'm wrong, during your explanation, you didn't tell us where opening sell/buy position is placed. Risk-reward is so important when trading, but you didn't mention that on this video. But, overall thanks (y)

[osazemeusen]

Impressive lecture, thanks for sharing

[kevinli]

Thank you for your incredible videos. Could you pleas let me know how I can possibly leverage this Synthetic data generation to apply to a panel dataset? Thanks!

[chrismoneystl]

15:35 stock prices are based on volatility

[BullBearInsights]

does this strategy work u=in indian market

[chrismoneystl]

1:20 more data, more test data, more back testing

[hbw]

How many data ( close prices) he uses on those models?

[chrismoneystl]

21:30 Medallion best years were high volatility.

[jordiplotnikovpous]

why do you take the logarithm of 1+percentage_increase, and not just 1+percentage_increase?

[lteodorescu]

so what's the actual strategy mate ?