iDov
-
Posts
121 -
Joined
-
Last visited
-
Days Won
3
Content Type
Profiles
Forums
Articles
Posts posted by iDov
-
-
"""
4h 50m | 881 MB
This course covers advances in the techniques developed for algorithmic trading and financial analysis based on the recent breakthroughs in machine learning. We leverage the classic techniques widely used and applied by financial data scientists to equip you with the necessary concepts and modern tools to reach a common ground with financial professionals and conquer your next interview.
By the end of the course, you will gain a solid understanding of financial terminology and methodology and hands-on experience in designing and building financial machine learning models. You will be able to evaluate and validate different algorithmic trading strategies. We have a dedicated section to backtesting which is the holy grail of algorithmic trading and is an essential key to successful deployment of reliable algorithms.
What You Will Learn
You will learn about financial terminology and methodology and how to apply them
Get hands-on financial data structures and financial machine learning
Understand complex financial terminology and methodology in simple ways
Ensemble models and cross-validation for financial applications
Backtesting for models and strategies evaluation and validation
Apply your skills to real world cryptocurrency trading such as BitCoin and Ethereum
Putting machine learning into real world problems and derive solutions
"""
May not be quite as good as it sounds, coming from a tech publisher well known for rushed production, but could well contain some decent info. Kudos to the original uploader!
- Traderbeauty, ylidor and ⭐ al2008
-
3
-
If people can't prove their claims then I don't believe a word.
Sure mate, do whatever works for you, I wish you don't lose money. I, personally, would never leave my trade decisions to a computer to make.
Sensible, but when someone doesn't have anything to gain by claiming X, there's less need for scepticism.
Seems like we just have different 'trading personalities' - no problem. I hope it all goes well for you too!
-
Can you show me those various people who are successful using a 100% mechanical approach to the markets? Can they show their broker's account statements?
He/they not trying to sell systems, they have no reason to show anyone anything...
-
1) The mechanical system is lke a sinewave in a fixed frequency.
...
2) In all books read so far, I have never come across a successful trader who relies on code or mechanical system, otherwise it would be too easy and a lot of people would be rich already.
...
I just think that your approach is the approach of jumping between systems when they stop working, that is documented as the losing approach.
1) The problem you identify with mechanical systems is mostly due to them using fixed lookback windows. Machine learning systems allow use of multiple indicators, so the problem is less serious. (ie there are ways of identifying which lookback periods are useful, even if that's changing.)
2) I believe all the 'Turtle Traders' and a good percentage of Schwager's "Market Wizards" used mechanical methods.
3) I've mentioned several approaches recently, but that's not 'jumping between systems', I'm just trying to figure out which I can systematize with the idea of combining them as inputs to an ML model. That's a long term project; shorter term I've coded up simple mechanical strategies for VIX trading from "Easy Volatility Investing" (Tony Cooper); next stop (hopefully) is combining them to form a better performing system (via ML), for some people who were looking for safer ways to short vol.
Anyway, each to his/her own :)
-
@logicgate: people can have different perspectives.
Last time I checked the point of trading was... to make money. Why bother to "become a master", if there are profitable non-discretionary systems out there that can be back-tested in seconds, instead of hours poring over historical charts? I'd rather spend that time working on coding skills - which are also more broadly applicable - and may allow some "discretionary" systems to be converted into non-discretionary ones.
-
"""
Learn How To Backtest Your Best Trading Ideas In One Day - Guaranteed!
How would you like to have the ability to backtest your best trading ideas? And learn how to do this from one of the best in the industry?
Introducing Programming in Amibroker - Learn How to Backtest Your Best Trading Ideas in One Day
Here is What You Will Receive
You will start the course with minimal to no knowledge on how to backtest in AmiBroker. Six hours later you will.
1) Know how to program a strategy.
2) Know how to backtest and validate that strategy.
3) Learn how to improve upon that strategy.
4) And be able to receive the daily signals for that strategy.
This is knowledge which you will be able to apply to your trading for the rest of your life.
In this class you will start from the beginning and within hours have the skill to take one of our Strategy Guidebooks and program it in AmiBroker yourself. Plus, your personal code will generate the signals for the upcoming day!
Course Objectives
This course is designed for traders who want to learn how to use AmiBroker to create backtests and/or generate trading signals, but who have little or no familiarity with the AmiBroker language.
At the completion of this course, you will be able to:
- Create your own custom indicators, and add them to an AmiBroker chart.
- Backtest basic trading strategies in order to see which ones have edges, which ones do not, and take the strategies that have edges and make them better.
- Verify that your backtest results are correct.
- Generate trading signals for the upcoming trading day.
What's Included
- Six hours of online instruction. The course was recorded live and you will be able to download it to your computer to view at your convenience.
- Several break-out sessions where you will spend hands-on time with AmiBroker.
- AFL code templates that you can easily modify for your own needs.
- A free copy of the Strategy Guidebook "ConnorsRSI Selective Strategy for ETFs and Stocks", which we will use as the basis for our backtest. By the end of the class, you will be able to take this strategy, program it in AmiBroker yourself, and get the signals for the upcoming day.
Prerequisites
- A desire to learn basic AmiBroker programming.
- AmiBroker version 5.5 or later installed.
- A data source configured to work with AmiBroker
"""
Kudos to the original uploader!
(Note: If you want more or different info on AmiBroker Programming, both Bandy books "Quantitative * *" include examples in AFL)
- homey, Traderbeauty, chungchung and 1 other
-
4
-
@logicgate
"You gotta gather the knowledge, and more importantly TEST the knowledge"
Exactly... & I'd need to automate a test of pitchforks to trust them/trading strategies based on them. The pivots are all that matters for pitchforks, but "choosing the obvious pivots" isn't really clear enough to code up.
AFIAK there are two objective definitions of pivots:
- d-degree zigzag filter -- a d-degree high(low) is at least d points (/pips/%) above(below) the previous low(high), and is 'anchored' when price reverses d points (or whatever) down(up)
- n-range max/min - an n-bar high(low) is higher(lower) than n bars to the left and right
Has anyone tried something as simple as these with pitchforks?
(My guess is they don't work very well if applied in such a broad-brush way, and that the 'action' (AB) move needs to look 'impulsive', with wide bars, closes near top of range, small lower candle wicks, modest pullbacks, gaps in the trend directoin. Summarizing all these into one objective measure of impulsiveness seems quite tricky.)
-
Curious to know what people think of it; some of the section titles make it sound like it might have some contributions to offer re trading systems development. (Although the 'Combining Rules' bit sounds like it might be a similar approach to the one in "Beating the Financial Futures Market" by Art Collins)
-
There's one thing that bugs me about all these methods: afaik none of them provide objective or even particularly clear criteria for choosing the A, B & C points that define the pitchfork. Just "AB is an impulsive move, BC is a correction". (& I watched quite a few hours of Morge and an English guy; forgot his name at one stage.)
-
-
Discover the 12-steps process to build trading systems that work on the live market! (2015/2016)
http://systemdevelopmentmasterclass.com/
Module 0 - Introduction
Module 1 - Overview
Module 2 - Your Key Idea
Module 3 - Development Environment
Module 4 - Sanity Check
Module 5 - Stress Testing Key Idea
Module 6 - Environmental Filters
Module 7 - Baseline System
Module 8 - Independent Testing
Module 9 - Combining Rules
Module 10 - Out of Sample Testing
It looks like these links are ok:
https://rapidgator.net/folder/5017023/OPEN%20LINK%20DOWNLOAD.html
I'm maxxed out on those two filehosts; if they're paid only complain & I'll delete the post. Kudos to the original uploader :)
- logicgate, ⭐ insaneike, ⭐ val2004 and 3 others
-
6
-
I can only find his earlier book :(
-
You could also check out David Weis "Weis Wave" indicator:
http://indo-investasi.com/showthread.php/19124-REQ-Weis-Wave-Training-Webinar
It's a lot easier to follow than his book (Trades About to Happen)
Sadly both of these approaches depend heavily on volume, which isn't available (at least in a particularly reliable form) for forex. Some Demark indicators offer a method for forecasting trend exhaustion points without using volume. They're also totally objective. But I don't think they're as reliable.
If anyone knows other good (ideally non-volume dependent) methods of identifying trends/trend exhaustion early, I'm all ears! (or eyes :p)
-
I can't see your screenshots but "Stop and Make Money" by Richard Arms has some good information on this.
-
Guo, Xin; Lai, T. L.; Shek, Howard; Wong, Samuel Po-Shing, Chapman & Hall/CRC, 2017
http://gen.lib.rus.ec/book/index.php?md5=F5D03C3D452EAB8A5F98D141EF183A06
- ylidor, ⭐ mr12323, ⭐ chullankallan and 5 others
-
8
-
Some kinds of historical data are very hard to get. eg long runs on intraday data for anything, options data etc. Perhaps there should be a subforum for sharing these.
-
When I read about Wyckoff/VSA it makes sense. It's far from obvious how to express it as a set of rules, but I remember "Tradeguider" is supposed to offer Wyckoff-based indicators. So I'm wondering if anyone has ever seen rules for a systematic Wyckoff/VSA system that I could use as a basis for trying to code it up? (Doesn't need to be something that works well, just an example of how it could be done...)
-
Native Python Indicators:
https://github.com/kylejusticemagnuson/pyti
Accumulation/Distribution
Aroon
-Aroon Up
-Aroon Down
Average Directional Index
Average True Range
Average True Range Percent
Bollinger Bands
-Upper Bollinger Band
-Middle Bollinger Band
-Lower Bollinger Band
-Bandwidth
-Percent Bandwidth
-Range
-%B
Chaikin Money Flow
Chande Momentum Oscillator
Commodity Channel Index
Detrended Price Oscillator
Double Exponential Moving Average
Double Smoothed Stochastic
Exponential Moving Average
Hull Moving Average
Ichimoku Cloud
-TenkanSen
-KijunSen
-Chiku Span
-Senkou A
-Senkou B
Keltner Bands
-Bandwidth
-Center Band
-Upper Band
-Lower Band
Linear Weighted Moving Average
Momentum
Money Flow
Money Flow Index
Moving Average Convergence Divergence
Moving Average Envelope
-Upper Band
-Center Band
-Lower Band
Negative Directional Index (-DI)
Negative Directional Movement (-DM)
On Balance Volume
Positive Directional Index (+DI)
Positive Directional Movement (+DM)
Price Channels
-Upper Price Channel
-Lower Price Channel
Price Oscillator
Simple Moving Average
Smoothed Moving Average
Standard Deviation
Standard Variance
Stochastic
-%K
-%D
StochRSI
Rate of Change
Relative Strength Index
Triangular Moving Average
Triple Exponential Moving Average
True Range
Typical Price
Ultimate Oscillator
Vertical Horizontal Filter
Volatility
Volume Adjusted Moving Average
Volume Index
-Positive Volume Index
-Negative Volume Index
Volume Oscillator
Weighted Moving Average
Williams %R
-
If you can watch something in-browser, it gets downloaded... Not always in a very convenient form. But if that's a major pain, it can also be recorded with screencasting software.
-
agreed, torrents are much better to work with than any downloading site.
But expose your IP address....
-
Yup, you need to register for 4shared.
Temporary login with G+ & Facebook also seems to work fine usually...
-
Can you upload to mega.co.nz? please
Mentioning price & mail => he wants to sell, not share, them...
-
Maybe check here first?
-
If the password is of more than 6 chars, then it will take forever to use a brute force method so don't waste your time on this method.
In fact pdfcrack doesn't even recognise the PDF as protected :s
Tom Demark method
in Forex Clips & Movies
Posted
If you find the Demark-authored books difficult to understand, Jason Perl's "DeMark Indicators" may be clearer (& has good charts):
http://gen.lib.rus.ec/book/index.php?md5=9094B3AAC6F3C77B9D40C1F0ECF82189
(I don't know if it covers everything in the books Traderbeauty linked to though.)