Determinism - Can Newtonian Physics Predict the Future? - Chaos, Fractals and Financial Markets
HTML-код
- Опубликовано: 1 янв 2024
- Whether you're interested in financial markets, quantitative analysis, or unraveling the mysteries of the universe, in this episode we will ignite your curiosity and broaden your understanding.
🤔 Ever wondered why some phenomena, like the weather, economies, and financial markets, are so challenging to predict? And why do financial markets dramatically shift their behavior, sometimes with very little warning? Join us in this thought-provoking series as we delve into the complexities that go beyond the confines of traditional scientific and mathematical approaches.
In this episode, we explore the age-old conflict between determinism and chaos, tracing back to Isaac Newton's deterministic worldview. Newtonian determinism posited that with precise knowledge of all components and physical laws, the future could be predicted through mathematical equations - the birth of Laplace's demon.
Here we uncover the fundamental shift from reductionism to nonlinear relationships in chaos theory, which challenges the belief that breaking down complex systems into smaller components is the key to understanding.
This series is targeted at those interested in trying to improve their understanding of how the real world works, and what methods might be most useful for analyzing financial markets, and actually for any kind of quantitative analysis involving complex systems. By understanding chaos and fractals you can improve your understanding of where traditional methods might work well and where they might fail. Hopefully it'll get you thinking more deeply about your analysis methods, for whatever purpose you use them.
We'll start this journey by looking at Determinism.
CREDITS:
Artificial Intelligence/Machine Learning - Gerd Altmann from Pixabay
Digits video - motionstock from Pixabay
Particles video - Tomislav Jakupec from Pixabay
Demon and Crystal Ball - Video - Victor from Pixabay
Quantum states - Sbyrnes321, Public domain, CC0, via Wikimedia Commons
Ferdinand Schmutzer, Public domain, via Wikimedia Commons
Pendulum Video - Grant Muller from Pixabay
Math book Image - sandid from Pixabay
Math board Video - Ciprian Stancu from Pixabay
Pendulums - Grant Miller from Pixabay
Snowflake photo - Egor Kamelev from Pexels
Coast photo - Oliver Sjöström from Pexels
Accounting class - Canva AI drawing 
Наука
Woah, you're back! Awesome!
Great video! :)
Thanks Brumor!!
Thanks❤
Thank you very much for watching and your kind comment! ☺
Happy new year ! Keep your great work on !
I would want to know in which cases randomness is not a good approximation for a chaotic systems
Happy New Year to you too!! And thanks for the comment. The main problem with randomness is that it usually assumes independence between timesteps, so has no "memory" of what happened before. In chaos the future does depend on the past, so it is in a way more predictable, but only in a limited manner. I'll have to make some more videos on this!!
You have to watch Booksy’s fractal videos. I think he is deep deep into understanding stock fractals and structure.
Hey again Juanro. Thanks for pointing me at Booksy's channel. I like his very candid approach about what he is trying to achieve. His ideas about fractals don't gel that well with my understanding, which is more based on the ideas of Mandelbrot for financial markets, and the methods developed to understand the statistical properties of geophysical and complex systems data (like long-range dependencies seen in the autocorrelation function, leading to power laws, fat-tailed distributions and the like). That's fine though - he has his purpose and own ideas. It's an amazing thing that so many systems have fractal-like behaviour in their statistics, even when there's no connection between them! So what I need to get on to is making some videos about how these patterns arise and the dynamics behind that. I might drop Booksy a message at some time though - so thanks for pointing out his channel! Cheers!
Hi bro thanks 🙏
Thanks! Glad you enjoyed it
but if describing chaos doesn't have meaningful predictive value, it's just as bad as the pure random walk bozos?
Not quite ... it does have short-range predictive value and the past does affect the future. Conversely, models based on random processes assume each timestep is independent. So there are insights from chaos that can be helpful in understanding market behavior which go beyond what random walk models offer. I agree random walk models are a pretty poor approximation of reality :) but can have some uses as long as you understand their limitations. Thanks, and great comment!!