If all you intended to do is to translate the code, I think you have done an excellent job on that and your work is now completed. You have a very clear understanding of the conditional statements in the code now and have mastered the translation. Excellent work. I think all who are reading this discussion appreciate what you have done. It appears that a much of the other input from participants is related to other trading platforms that can do better than the indicator. That information is good to know, because some readers might like to change trading platforms.
I only make reference to TradeGuider's publications, because Williams is the only author that I've found that has quantified VSA to the point where it can be programmed. If there were other publications that are as detailed has his, I would like to read them too. What is the ultimate goal of an indicator, if it is not to help the trader to know the best times to trade. Perhaps, in time, when I have a deeper understanding of VSA and of the code in Karthik's indicator; I can code the reversals myself. I'm not at that point yet. I've contributed what code I can with the understanding of thinkscript that I currently have. I didn't realize that reversal signals were bells and whistles, but I respect your opinion. Perhaps, I misunderstood what you want for user feedback. If you like, I can critique the timing of the symbols relative to their definitions. As for the three time frames, no one seems to know the reasoning behind the formulas; so they are nearly impossible to evaluate. I've tried to find Karthik's discussion of his indicator on the Traders Laboratory Forum, but I get an error message "The discussion thread you are trying to reach has been removed". So we don't know if his code has had a peer review like yours. For now I think your translation of Karthik's code is complete and anything beyond that are improvements beyond the original code. You've done an excellent translation.
Last edited by StockJock; October 9th, 2010 at 11:45 AM.
The following user says Thank You to StockJock for this post:
Thanks, but I didn't do the translation - cbritton did. What I did was fix a few bugs, change a few symbols and arrange them in a logical order based on strength of signal, change the color scheme from one that was totally random to one that makes logical sense, and add a few enhancements. All of those changes are detailed in previous posts and are marked by C_RP comments in the code itself.
So far, it's not clear to me that TG is consistently better than this study. In addition, there's a slight price difference between the two. TG's former list price was $5K and is slated to be leased yearly at a rate of (I believe) $2K. Our VPA study is, as Harvey Levin likes to say on TMZ, Free!
This is exactly the kind of specific information that would be helpful in tweaking the study so that it functions optimally. In previous posts I have mentioned things that seem to work pretty well and other things that don't work at all. But frankly I've never seen any indicator or study that works 100% of the time.
I agree that the time frames issue should be addressed. Any info that you can dig up on this would be appreciated.
Thanks for your input!
Last edited by snowcloud; October 10th, 2010 at 07:52 PM.
The following user says Thank You to snowcloud for this post:
Here's an example of how the study works better when volume is more robust. Notice how the symbols before 5:45, while not totally meaningless, do not provide particularly strong indications of direction. Just after 5:45 when volume begins to pick up the symbols are much more clear and give quite accurate indications of future price movement. Even after 6:45 the symbols provide accurate indications of direction even though the movement in those directions is not very strong.
(Btw, the two horizontal purple lines are former support and resistance levels, the two dotted green lines are Pivot Point support lines, and the bracketed lines are a Fibonnaci retracement).
As I think about the challenges facing traders today, I keep coming back to two fundamental truths: First, the result of institutional activity is revealed by volume and price spread. Second, significant turns in the market take time to orchestrate (accumulation and distribution).
Today's market has continued to rise (in the face of many conventional over-bought and negatively divergent signals) ...largely because liquidity inflows have been expanding. There are 3 possible sources: 1) the Fed, 2) foreign investment entities or central banks, and 3) large Wall Street firms such as GS.
In a market, where conventional indicators (MA crossovers, etc.) are of diminished value and the sources of liquidity are camouflaged, the ability to distinguish commercial investment activity from short-term speculation and to recognize the differences between strength and weakness is critical.
That is what Volume-Spread-Analysis is about. And despite the algobots' short-term manipulation of liquidity, the broad markets remain limited to four phases: accumulation, markup, distribution, markdown.
The point: I think VSA is more important now than ever before.
Last week, I compared 3 systems: 1) an aggregate algorithmic system, 2) VSA (principles-based, not TradeGuider), and 3) Auction Market Theory. I analyzed 6 months on the ES. The result: There were times when significant market turns took place and only one system signaled. There were many times when 2 of the 3 signaled. But clearly, the most precise, the most anticipatory, and the most reliable system was the application of Wyckoff-oriented principles. They work fine today and TOS is as good as any other platform at providing an interface to color code common setups.
Key point, Snow. The greater the liquidity, the more difficult it is to manipulate the market. Moreover, the more time it takes to effect change.
I would urge those interested in this thread to look at the most liquid markets - and after getting an overview (monthly, weekly time-frames) ...look at how nice the signals are in the 4-Hour time-frame. Once you see the first sign of a trading signal, look inside that bar with the 2-Hour/1-Hour time-frame.
Using these time-frames, a lot of the noise and emotion is absorbed. Excellent trades appear in these time-frames; no need to fear missing out when some news event provides catalyst for what is often already set up. Significant moves only occur on volume.
Example: Worried you missed the rally in Silver? On Friday, a steep decline was followed by absorption volume and then a low-volume retest of the area buyers came in. A temporary supply shortage provided the setting for buying low volume reactions. Standard VSA stuff. These occur regularly, in both directions - supply and demand-side.
Liquidity and time-frames that themselves filter out noise - key.
The following 2 users say Thank You to swimtrader for this post:
I don't know how it could be much more clear than this. Check out the two Stopping Volume symbols on the attached 6e chart. The first is at the end of the Ben Bernanke speech pop. The second is at the bottom of a long downtrend which led to a 30 pip pop up before turning down again [continuation of downtrend not seen on chart]. The very brief and shallow up move following the Green Circle Stopping Volume 10 bars before the labeled one is less dramatic but still valid, good for about 12 pips or so.