All Things Statistical - page 5
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Proof For Support & Resistance Levels
Support and Resistance Levels collected from multiple firms are collected and subjected to Bootstrap testing. The arbitrary levels are found to have prices bouncing off them 56% on average, with results statistically significant at the 5% level. These intraday price levels are found to last for at least a week too and are robust to changes in testing methodology.
Frankly, this seems like the 1st ever formal/statistical/objective study ever done on Support and Resistance lines. Enjoy!
Wintersky
0007osle.pdf
Open High Low Close/ Range Based Volatility Estimation
In addition to Garman and Klass's 1980 work on extreme range volatility estimation, the paper below states that the usage of Open and Lows of prices commonly used in Technical Analysis has better efficiency than just close-to-close volatility estimation.
http://fbemoodle.emu.edu.tr/pluginfile.php/40034/mod_resource/content/1/Towards_the_fundamentals_of_Technical_Analysis_1_.pdf
Another crucial backing of the supporting of high and lows would be in the usage of Internal Bars Strength (IBS), which has the same formula as that of a simple TA stochastics indicator. Though it is used in the Quant sphere for machine learning due to the scale invariant and stationary properties, the emphasis of High and Lows again suggests perhaps that relative price ranges/volatility would offer a crucial edge in financial spheres.
Intelligent Trading: IBS reversion edge with QuantShare
Wintersky
Interesting
An interesting thread at FF: "Use Of Statistics In Trading" which actually touches more on Digital Filtering aspects. A very enlightening read.
Can statistical methods be used to create a trading strategy? @ Forex Factory
Wintersky
Reasons For Doubting Regression
An elementary article on the problems with Regression Assumptions. Simple certainly, but good for raising one's uncertainty level and checking the assumptions behind all indicators/model constructs which as we all know, are rarely fulfilled in real life
Regression Fantasies: Part I | Stats With Cats Blog
Wintersky
This is one good thread
Thanks man
This is one good thread Thanks man
Welcome. All im doing is posting things i find interesting. Though im still learning and cant be absolutely certain if they are indeed worth the hidden value i find them to be worth. So it would be good if anyone could add any discussion of sorts or provide some interesting reads
Wintersky
R-Squared
An Elementary introduction to R-Squared which i chanced upon as i was reviewing the statistical tools available at our disposal.....
TASC article on R-Squared:
identifying_market_trends_1.pdf
A Visual Simplified Explaination of R-Squared:
R-Squared: Sometimes, a Square is just a Square
Proper Interpretation of R-Squared:
Regression Analysis: How Do I Interpret R-squared and Assess the Goodness-of-Fit? | Minitab
Why NOT to use R-Squared for Nonlinear Regression:
Why Is There No R-Squared for Nonlinear Regression?
Enjoy
Wintersky
Kalman Filter
Good Day All,
Below are some of the papers on Kalman filter that i find to be quite user-friendly for the not-too-mathematical population like me.
http://www.cl.cam.ac.uk/~rmf25/papers/Understanding%20the%20Basis%20of%20the%20Kalman%20Filter.pdf
Bilgin's Blog | Kalman Filter for Dummies
http://www.cs.unc.edu/~welch/media/pdf/kalman_intro.pdf
====================================================================
I find Kalman filter to be very interesting/thought-provoking in many ways. If there's something good about the Kalman filter, it seems to be the fact that given fixed assumptions, especially about the mean, it is extremely useful. Hence, it is not difficult to see its popularity in alot applications in electrical engineering, physical problems etc.
The pictures here both come from the PDF in the 3rd link i posted. What makes this fairly interesting to me is the fact that the convergence properties are very good frankly, converging very fast after 50 iterations.
Under noisy measurements, convergence still works fairly well. Yes, the biggest beef we have here is the changing mean characteristics in the actual world. We cannot really be certain if the first few moments really exist (mean, variance, skewness, kurtosis etc). One way i can see this being useful is if we use if at the higher timeframes. Or being extreme and going down to the very low timeframes, letting there be more time for convergence under changing mean conditions. But frankly, i would think that this indicator needs to be adapted in some way to make it more useful.
I dont know. i have heard that some people around have made good systems inclusive of a Kalman filter too. Feel free to chip in your views, if any.
Wintersky
Cheers
Theoretical Properties Of Range
Another rare article supporting the use of ranges as partial justification for the validity of Japanese Candlesticks. So we have now a total of 2 articles regarding ranges. Enjoy!
http://omicsgroup.org/journals/theoretical-properties-of-technical-range-and-its-applications-2168-9458-1-103.pdf
Wintersky
Cheers
Honouring John Tukey
............................................................