Yoga is better.
Do you see patterns in Random Walks?
- Thread starter atlTrader666
- Start date
Yoga is better.
Okay, Morpheus
I have been thinking a lot lately about Thomas Bayes and his remarkable discovery. It is hard to believe that someone in 1700s could have such a vision! It has been forgotten for over 200 years and now there is even a brand new branch of Artificial Intelligence that is based entirely on Bayes's Theorem. Since I have been implementing this approach lately in my own work quite successfully I have decided to dig a bit more in this direction. It is quite fascinating! I also have realized that there is no good portrait of Reverend Thomas Bayes. So, I have decided to create my own. It took me about a month and a lot of reference images to create it, but I am glad I finally finished it. I used to paint a lot, however in recent years I was too busy and did not have time for art; not good! I am glad that I took time to make this portrait. Please let me know what you think.
Cheers,
MAESTRO
P.S. According to a research I found online 97% of all the people who were exposed to Bayes's Theorem still do not understand it entirely and cannot implement it in their work properly.
Cheers,
MAESTRO
P.S. According to a research I found online 97% of all the people who were exposed to Bayes's Theorem still do not understand it entirely and cannot implement it in their work properly.
Then you are aware that Asimov may have been a psychohistorian himself, and we are in the final stages of the "Galactic Empire". That said, how can traders, most of whom sense something is amiss without knowing exactly what, benefit from what you are trying to convey?
In other words, what do we need to know, and where do we start?
Huh... Baye's Theorem...?! Well now you've given me something to think about. I haven't read the entire thread... Do you apply this to trading...?
Quote from MAESTRO:
I have been thinking a lot lately about Thomas Bayes and his remarkable discovery. It is hard to believe that someone in 1700s could have such a vision! It has been forgotten for over 200 years and now there is even a brand new branch of Artificial Intelligence that is based entirely on Bayes's Theorem. Since I have been implementing this approach lately in my own work quite successfully I have decided to dig a bit more in this direction. It is quite fascinating! I also have realized that there is no good portrait of Reverend Thomas Bayes. So, I have decided to create my own. It took me about a month and a lot of reference images to create it, but I am glad I finally finished it. I used to paint a lot, however in recent years I was too busy and did not have time for art; not good! I am glad that I took time to make this portrait. Please let me know what you think.
Cheers,
MAESTRO
P.S. According to a research I found online 97% of all the people who were exposed to Bayes's Theorem still do not understand it entirely and cannot implement it in their work properly.
Good portrait...
Did you paint that? That's really good, good to have a hobby outside the computer.
I read about Bayes theorem in my probability book (http://www.dartmouth.edu/~chance/teaching_aids/books_articles/probability_book/book.html, they put on GNU license so anyone can download and study). The examples I've seen in the book and a couple other sources, they always use the example of breast cancer. The formal definition trips me up. Here's my simplified template from the example:
1) there is an actual percentage of the population who have it
2) the test has an error of some amount
3) of those who get a positive result, take out the % of: (actual % population that does not have it * test error)
4) this gives you: of those who tested positive, how many actually have it
So, even if you get a positive test result, it's a less than one probability, and the probability is reduced more than just the test error rate, but the number of results that are false positives.
I haven't found people to discuss it with to learn more interactively, like you say most people haven't been exposed to or understand it.
I read about Bayes theorem in my probability book (http://www.dartmouth.edu/~chance/teaching_aids/books_articles/probability_book/book.html, they put on GNU license so anyone can download and study). The examples I've seen in the book and a couple other sources, they always use the example of breast cancer. The formal definition trips me up. Here's my simplified template from the example:
1) there is an actual percentage of the population who have it
2) the test has an error of some amount
3) of those who get a positive result, take out the % of: (actual % population that does not have it * test error)
4) this gives you: of those who tested positive, how many actually have it
So, even if you get a positive test result, it's a less than one probability, and the probability is reduced more than just the test error rate, but the number of results that are false positives.
I haven't found people to discuss it with to learn more interactively, like you say most people haven't been exposed to or understand it.
Old but funny. Somewhat related to this topic and machines that were mentioned in this and HFT thread.
http://youtu.be/UtAwqyfCmDw
http://youtu.be/UtAwqyfCmDw