1,151 reputation
38
bio website
location
age
visits member for 3 years, 8 months
seen Sep 27 at 14:47

May
19
comment Source of Quandl Open Data
I'd like to know this "straight answer."
Mar
5
comment How to identify technical analysis chart patterns algorithmically?
I second the premise of this answer. I think that applying machine learning to pattern recognition is a well established field.
Jan
3
comment How can I go about applying machine learning algorithms to stock markets?
This is much the same as the approach I'm taking with Neural Nets and blogging about at dekalogblog.blogspot.com I would be interested to hear more about how you're getting on with your version of the same.
Sep
17
comment How should I include the bid-ask spread as a transaction cost in a backtest?
After receiving a signal from daily close data the earliest opportunity to open a trade would be the following day's open. Why is it "common practice" to use the following day's close? This implies that one waits a whole trading session before acting on the signal.
Apr
25
comment Why in general is the variance of volume changes higher than variance of price changes?
Accepting that this is true would the implication be that indicators that combine price and volume, such as OBV, money flow index etc. are fundamentally flawed?
Dec
20
comment Time series price prediction and linear regression: using high/low rather than last quotes price
@TalFishman Could you supply a reference for the approximate VWAP formula you give in this answer?
Nov
29
comment How to generate a random price series with a specified range and correlation with an actual price?
This is such an intriguing approach that I may "knock up" some Octave code on my blog and post a link to it from here.
Nov
29
comment How to generate a random price series with a specified range and correlation with an actual price?
Interesting link! I theorise that one way it could be done is to apply a Fast Fourier Transform (e.g. FFT function in Octave/MATLAB) to the original series, then divide the resulting vector into segments and within each separate segment randomly permute the values, and finally reconstruct the time series by applying the Inverse Fast Fourier Transform (IFFT function). By randomly permuting the values you will not alter the amplitudes of the composite sine waves but will slightly alter their periodicity - this will naturally tend to restrict the range of the new series to that of the original.
Oct
4
comment How do I calculate expectancy from a past series of trades in my trading account?
Another way to recast the trades is in Van Tharp's concept of R multiples, e.g. 2R, 0.5R, 2R and 2R, the average of which is 1.625, i.e. on average you can expect to make 1.625 times your risk per trade.
Oct
2
comment What should be considered when selecting a windowing function when smoothing a time series?
Please ignore my last comment. I tried to post the link to the answer at dsp.SE as an answer, as per Tal Fishman's comment, but the forum software automatically converted this to a comment.
Oct
2
comment What should be considered when selecting a windowing function when smoothing a time series?
This question has been answered at dsp.SE [here][1]. [1]: dsp.stackexchange.com/questions/208/…
Sep
18
comment DSP: stationary non-periodic signal: what's the best causal technique?
@Bootvis Code for the JMA is available here link
Sep
15
comment What should be considered when selecting a windowing function when smoothing a time series?
This post has been answered at dsp.SE [here][1]. [1]: dsp.stackexchange.com/questions/208/…
Sep
12
comment What should be considered when selecting a windowing function when smoothing a time series?
@sheegaon - thanks for the heads up about dsp.SE. I have now posted this question there too.
Aug
16
comment How to generate synthetic FX data for backtesting?
I agree that there are some limitations and I have posted my adapted version here
Jul
20
comment Free paper trading site with an API
I've only used manual entry on their API for paper trading.
Apr
13
comment How does return-based analysis calculate expected return of a trading system?
@RockScience. The tests are a Monte Carlo permutation test and a bootstrap test on detrended returns.
Apr
12
comment How does return-based analysis calculate expected return of a trading system?
With regard to concerns over data mining, it would be trivially simple to apply the tests outlined in Aronson's book, "Evidence Based Technical Analysis," to the above posited system to determine the extent to which data mining bias has an influence on the system's returns.
Apr
3
comment Trading C++ Libraries
The OP wanted "some of the functions that would be used in developing a trading strategy." Although I cannot quote any evidence in support, I am pretty sure that the tools of technical analysis are/can be used in developing such strategies. As to whether TAlib is written in C or C++, well I stand corrected.
Feb
8
comment What concepts are the most dangerous ones in quantitative finance work?
I remember reading somewhere a comment that seeking non correlation is a complete waste of time; you want all your strategies to be 100 % correlated to a exponentially rising equity curve.