351 reputation
15
bio website
location
age
visits member for 2 years, 2 months
seen Oct 4 at 2:19

May
13
comment How to design a custom equity backtester?
Amibroker is great. There are also several good books written by Howard Bandy, to gain practical experience with the language.
Oct
13
comment Practical Usage of Wavelets with Real Time Data
The issue is pretty well known and a good discussion can be found in amazon.com/Neural-Novel-Hybrid-Algorithms-Prediction/dp/…. In the text, he discusses boundary effects and transformations to mitigate them.
Jan
30
comment Predict Quadratic Trend in Time Series
agree w/richard. The code should not even run without errors as it is displayed.
Dec
30
comment Fastest algorithm for calculating retrospective maximum drawdown
There is only 1 path from inception (and only 1 iteration required with vectorized dd result of that path). What you are describing above is a rolling dd (*Note I specified since inception). If you have a time series and can show an example of your algorithm and time, I will reproduce using a vectorized approach and compare times.
Dec
30
comment Fastest algorithm for calculating retrospective maximum drawdown
A vectorized approach (since inception) is very fast.
Dec
25
comment Are there any tools or useful algos for identifying corner portfolios?
*Also, see zivot links below. "The the set of efficient portfolios of risky assets can be computed as a convex combination of any two efficient portfolios."
Dec
5
comment Why are regressors squared and not ^1.5 or ^2.2 or ^2.5?
Because a negative number raised to a non-integer exponent is complex. Of what use would regressing physical economic values have here?
Nov
29
comment Bootstrapping first, then data mine?
I have heard good things about meboot, and started to run it, but I'm not so sure it captures both auto and cross correlations well. It doesn't really make any sense to do cross validation after the data set has been mined. The ideal method would be to train/validate/test over each slice fold set... so that every slice is not biased in any way by the slice that is left out. If you have a strategy, however, you could run the CV over parameters of the same strategy to evaluate robustness.
Nov
20
comment When gains are made: Overnight or during trading hours? What is the connection to volatility?
Paying Attention: Overnight Returns and the Hidden Cost of Buying at the Open -- papers.ssrn.com/sol3/papers.cfm?abstract_id=1625495
Nov
7
comment Generate tick data from candlestick
It would be akin to extrapolating daily price data from only having a yearly data OHLC source. One might be better off by at least generating bootstrapped data from a some samples of tick data. At least then, some of the properties might be captured.
Nov
1
comment Coin Toss System
@Freddy. You stated, "I have not heard of long-term successful "market operators" to generate profits off the back of predicting what the next tick/bar/day is gonna be as a function of the past couple observations, dictated by certain observed or quantified "patterns". " I gave one concrete illustration where that was the case. I agree with just about everything else.
Oct
31
comment Coin Toss System
I'll throw a disagreement (only) on the pattern side (although I do agree few approach it successfully). As an example, please see Toby Crabel's work and accomplishments (he utilizes the conditional probabilities of simple 'coin toss' like experiments as a pattern based decision). There are other more complex ways to apply the logic as well. But the key, as you point out (IMO), is that markets have properties like drift and momentum that make such methods feasible
Oct
31
comment What are the proper metrics to look at for checking discrepancies in these two time series
It would be interesting to see a scatterplot of the data. Even though you have poor cross correlations, you might be able to detect some level shifts in the plot and formulate if there is an easy calibration for the shifts.