![Noah Silverman](/img/default-banner.jpg)
- 12
- 132 722
Noah Silverman
Hong Kong
Registrace 18. 04. 2006
Helios.ai
We transform complex data into actionable business strategy using advanced statistics and machine learning.
"Arbitrage through information"
We transform complex data into actionable business strategy using advanced statistics and machine learning.
"Arbitrage through information"
Blockchain Mechanics @ Hong Kong University
A general, and introductory discussion of blockchain concepts.
zhlédnutí: 317
Video
Short clip on state at the Asian Financial Forum. Hong Kong.
zhlédnutí 308Před 6 lety
Short clip on state at the Asian Financial Forum. Hong Kong. Asian Financial Forum | 2018亚洲金融论坛
Startup Launchpad HK 2016
zhlédnutí 487Před 7 lety
I discuss machine learning, AI, and data science for startups.
TEDx 2016 - Privacy and Advertising
zhlédnutí 249Před 7 lety
I discuss the intersection of online privacy and the advertising world. Talk given at TEDx Hong Kong on July 15, 2016
Probabilistic Programming and Healthcare
zhlédnutí 594Před 8 lety
Short talk given at Metta in Hong Kong. I discuss some general concepts of probabilistic programming, and demonstrate a healthcare motivated example.
Artificial Intelligence - Under the Hood
zhlédnutí 853Před 8 lety
An overview of the current state of Artificial Intelligence, some current trends, and a very brief introduction to machine learning models, and more. Presented at the KPMG Change Face of E-Commerce event, June 2016 in Hong Kong My talk starts at 1:00.
Introduction to Artificial Intelligence at Metta Hong Kong
zhlédnutí 410Před 8 lety
Introduction to Artificial Intelligence at Metta Hong Kong
Bitcoin Orderbooks and High Frequency Market Microstructure
zhlédnutí 23KPřed 10 lety
Originally presented at Inside Bitcoin Hong Kong in July 2014. www.smartmediacorp.com Bitcoin Orderbooks and High Frequency Market Microstructure. A study of how Bitcoin order books behave, and how size of queue effects prices. I discuss modeling the waiting time until a price change, and the probability of an upward price move.
Bitcoin Trading: A Statistical Analysis
zhlédnutí 3,6KPřed 10 lety
www.smartmediacorp.com A quantitative analysis of Bitcoin price swings and market behavior. Includes discussion of using a properly tuned MACD as a possible trading strategy.
Predicting Horse Race Winners Using Advanced Statistical Methods
zhlédnutí 102KPřed 10 lety
Conditional Logistic Regression with Frailty applied to predicting horse race winners in Hong Kong. www.helios.ai Since first proposed by Bill Benter in 1994, the Conditional Logistic Regression has been an extremely popular tool for estimating the probability of horses winning a race. I propose a new prediction process that is composed of two innovations to the common CLR model and a unique go...
I think better to take questions at the end.
Investing with Mrs Vera has been the best step I have ever took in my life, may God bless that day I meet her
Good explanation of such complex topic. 12:20 - The 2D chart is interesting...
Excellent video. Have you considered implementing reinforcement learning with your data and inputs to make recommendations?
great lecture, would definitely be interested in the more maths related you alluded to at the start
I wish I could buy 1 bitcoin for $99
Great content. Thanks you so much for sharing!
Ooh
Your suggestions of Maker/Taker fees aged nicely, as well as the streaming order book.
Thank You!
Really interesting talk! At around the 21 minute mark you mention an article on your blog. After googling around it seems that your blog was published on btcmath.com, but unfortunately the website doesn't exist anymore. Any other place I can access your old blog posts?
great video, some of it went over my head. interesting to see the market has evolved like you said it should.
Glad you enjoyed it
Is this only optimal at Hong Kong could this be used a Fonner Park in Nebraska?
Dear Noah, thanks for the insights and analysis. Is there a way you would share the Python code you mentioned and used for your analysis? I would love to look at the data now in 2020 and use this analytical framework.
Sorry, that code isn't avaialbe for the public
at 4:05 if I remember my 8th-grade math correctly does ∝ mean that there is a constant in the formula or am I an idiot?
Interesting :) But i guess iceberg orders would cause you problems if your time window is too big. Lots of improvements have been done since 2014 when you recorded this video. Wish i found this sooner :)) Thanks. Have a nice day!
Could you elaborate on what improvements? Thanks,
The problem we , all the gamblers face in the end is: how race is going to develop knowing that the betting companies know prior to the start 1.) The amount of bets placed on a particular horse and 2.) The value of this bets.. Basically every thing can be controlled (manipulated) for the benefit of betting companies, otherwise this companies will cease to exist. If I misspelled something then please Pardon my French, but English is not my first speaking language
by the way, benter had hired journalists so they could get him some insider info.
Would you kindly provide a new clip of the demo ?
Hi Noah , have you tried to use or reconstruct these order books ? If so what are the best algorithms in your opinion to do so , does MC type of estimators / algos would be effective ?
this was way before its time. Thanks for the great upload!!
Please keep uploading new videos. The stuff you discuss is very valuable.
I need more information about econométrica method and betting Sports, please
could an exchange stagger the arrival rates and throw all of this off? Or not really.. because the market efficiency requirements dictate that they need to happen so fast, that by polling at longer intervals they average out and balance anyway?
They could, but I don't see why they would bother. Exchanges generally want to handle orderbook messages as fast as possible, and in strict time priority. Anything else would be considered "manipulation", which would be unethical, and probably illegal in any regulated market.
@@NoahSilverman Right. That's what I was thinking. Only reason would be if the predictability in programming from being so new..produced an inefficiency in the market. But not really an ethical way around it, because like you said.. The speed is already so high.
what would you call the VWAP that has a greater depth into the order book? Right before you talked about gasp.
waiting time and directional move looks golden. gaps are the key. this is cool stuff. thank you for sharing.
Winner of a video, I have been researching "horse betting nyc" for a while now, and I think this has helped. Have you heard people talk about - Senameron Sanyatt Approach - (do a google search ) ? Ive heard some amazing things about it and my mate got cool results with it.
This is all nice, but tell me which horse is going to win the first race at Aqueduct tomorrow.
Noah, Got two questions for you; 1) Does using conditional probability give you any advantage over just using the probability and ranking the horses grouped by race? 2) Are you able to provide any detail on the features that you've used? I'm looking at doing something similar for my MSc dissertation. thanks, Matt
What about hidden orders?
Soo.... Have you made your billions yet?
@3:38 when you say a seller wants to fill his 1 BTC sell immediately so he goes to the buy side and takes out a buyer, does the price of BTC then fall to $99 since the last transaction happened at 99? second question: if a buyer comes in and takes out 2+10+1 btc @100 how does the buy side fill up? i see on coinbase pro immediately buy orders come in at 100, where does this come from so fast?
Five years old but still an interesting talk. I thought about adding some thoughts on a few points but I realise there are so many different ways to predict races that my thoughts probably won't make a difference. Bill Benter's story (and his associates) is pretty far out considering the technology of the time when he was doing his thing with horse racing. Maybe he still is betting along with his academic pursuits, I don't know. Personally I bet on horses every day using my own spreadsheet formulas but the latest thing is with the help of a data science guy we are developing a deep supervised-learning ann using the parameters I know work best for the data sets. It's working somewhat but time will tell how well that goes. As for ROI, accuracy of prediction and staking methods, I believe that when someone is getting tangible results over a period of time they probably won't be telling others how it's done. Even if they do I remember something Bill Benter said which really resonated with me and that was many people don't want to roll up their sleeves and do the hard work. The countless hours I have put into writing formulas or making data sets, lol, I don't even want to think about it. Anyway, it's all fun :)
Super work , I am a Chinese and it's very interest in Hong Kong racing research. How I can learn that .and using your data for it
Can you provide the link of the research paper/journal you followed?
How can I calculate GASP?
Here gist.github.com/iUmarov/ff3689dcbd93ac081e3c4f526caa8a4a
great video thanks. Are you running a trading bot in bitcoin market?
Hi Noah, The video was really enlightening but can you provide link for the paper so that we can develop a decision model for research purpose. I am unable to connect few missing dots like how to identify new orders, cancelled order or filled order from the order book data as it is just order book at some instance. Pardon me if I sound too naive. Thanks in advance.
There is no paper available. Up to you to figure out how you want to implement.
I agree with Bryce, and thank you for sharing. The bot info was an eye opener. I new it existed but you graphically displayed it. Would love to see plain vanilla options too and the ability to buy or sell premium with any token with automated conversion.
This was dope. Extremely easy to understand. Thank you.
Thanks Bryce. Glad you enjoyed it.
Thanks
Unless someone has inside info about a race, there is no reliable way of predicting the outcome of a thoroughbred horse race. There are too many variables, not the least of which is the horse itself, whose temperament and condition at post time is known only to the horse, and the horse is keeping that a secret. The fact that even the most successful jockeys win only a small fraction of their races is proof that, presuming that the races are legitimate, the outcome is not a sure bet. Recently, in a maiden claiming race, the 75 to 1 longshot won by two lengths while the 6 to 5 favorite came in eighth. Predicting races is entertaining, but don't expect the horses to cooperate. They have other concerns that have nothing to do with money.
I respectfully disagree (of course)
Tell Bill Benter that there's no reliable way of predicting the outcome of a horse race lol
"Predicting races is entertaining, but don't expect the horses to cooperate. They have other concerns that have nothing to do with money." LOL FOFL My coffee nearly came out of my nose. GOOD ONE
You would have to use Kelly and the law of large numbers to mitigate uncertainty and bad luck. Is that what you would do Noah?
The reason why people can win money in Hong Kong field is that the pool is a pari-mutual pool with many punters without intelligent that there are rooms of different between the probability and odds
Judging from the video description the betting public underestimates the winning chance of a horse in 2 out of 10 races in Hong Kong, enough to overcome the track takeout over the long term. Is that correct? What betting strategy was simulated? Flat betting the bare minimum or a fixed proportion of the bankroll? This is important to know because horse racing typically doesn't encourage the implementation of a Kelly Strategy with a large bankroll relative to the size of the parimutuel pool.
Mr. Silverman, thanks for your talk. I'm wondering if you can give any suggestions on possible sources to turn to for a novice without a statistical background and wants to bet on horses using statistically proven methods. Thank you.
Really enjoyed this talk. Thanks very much for posting it.
Thanks Gene!
Really interesting, I've been seriously looking at bitcoin for less than a year, but it's clear that much of your "wishlist" at the end of this vid has been fulfilled. Are you still paying attention to bitcoin? Would be interested to hear an update of your thoughts!
I do still work with Bitcoin - much has improved since that talk.
I am an associate professor for a financial engineering major in bachelors of engineering, this semester i will lecture "Trading Systems and Market Microstructure", i am really interesting in your material and your experience, do you have any opensource project ? or willing to collaborate in an common license schema to develop applications and educational material for crytpocurrencies and trading with market microstructure focus ? . I am working with a real time orderbook querying and your framework it is very very useful. Are you interested in allowing me to use some of your material and contribute with working examples ?. please write to if.francisco.me@gmail.com
What will happen if a quantum computer give you a optimized result in fraction of second, and ruined the whole industry?
Nice fantasy, but things don't work that way. Just because a machine is "quantum" doesn't mean it has infinite insight into any phenomenon in the world.
but if you have a model, then all the way out is get a optimized answer, which i think the quantum machine D-wave in google can do the rest of answer, isn't it?
are you saying you would combine the public's implied odds (strength) with your coefficients? you're using public odds as a coefficient?
Lot of racing models use the public odds as *one* of several factors. There is information in there.
And, to clarify: We have "factors" in the model, and then use machine learning techniques to estimate the coefficients (weights of the factors). So, the public odds is a "factor" not a coefficient
so this differs from benter slightly as he suggested running a second logit model with combined public estimate and your fundamental estimate?
There are many ways to do this.
Noah Silverman last question, do you think your model's rsquare outperforming public model rsquare is a good indicator of potential success (along with OOS testing for ROI)?
Hey Noah, excellent talk. How did you get the .3 to .4 correlation between the odds and rank outcomes? Is that a number that you computed or something that comes from the academic literature? if you could provide a reference I'd be very grateful. Thanks.
Empirical correlation from dataset. If you want a formal "academic literature" reference, see my paper published on the topic.
Thanks for a great overview of your modeling. Are you using any open source libraries to do your conditional logistic regression and the LASSO optimization? Did you write this in C++ for the MAC? Thanks for any information you can provide on your algorithms.
This project was all custom code. Some in R and some in C++