StockFetcher Forums · Filter Exchange · BUYING THE DIP<< 1 ... 4 5 6 7 8 >>Post Follow-up
dknoonan
27 posts
msg #115552
Ignore dknoonan
9/26/2013 4:44:52 PM

Yes. It is one of the many things I've tried to improve results. Over most time spans, close above open (a white candle) gets worse results. I was figuring that entering at ROC(7) < -2 means a descent, so why not wait until it starts its ascent before you get into the trade? But I think that getting into the trade after the first white candle has passed means I miss out on the gain of the first white candle. In most time spans this seems to greatly reduce the gains of the filter. So I commented it out. A similar effect occured when I put "high 1 day ago" as the entry price, and set Conditional Entry to Yes.

So, using this filter to trade is like the old phrase, trying to catch a falling knife. Funny thing is, it catches it pretty well. Or at least it goes up after falling a couple of days after your buy.

Kevin_in_GA
4,552 posts
msg #115557
Ignore Kevin_in_GA
9/26/2013 6:51:02 PM

One could always use a limit order, but you might miss out on a bunch of good trades. Avoiding drawdown is important but for me frequency is what needs to be maximized if the system gives you a positive expectancy.

Note that the last 4 months have not been kind to this system, generating about a 3-4% loss. All part of the larger landscape.

jimmyjazz
99 posts
msg #115571
Ignore jimmyjazz
9/27/2013 10:00:48 AM

Kevin, how does this system different from your z-score approach?

Kevin_in_GA
4,552 posts
msg #115581
Ignore Kevin_in_GA
9/27/2013 12:27:33 PM

The z-score looks at the relative relationship between the stock and the ^SPX. This only looks at ROC.

dknoonan
27 posts
msg #115668
Ignore dknoonan
10/2/2013 1:15:16 AM

On 6/3/13 Mahkoh said this about doubling down to recoup some of the drawdowns:

"In my example from 5/20: Originally 282 trades from 5/18/2012 until 5/18/2013. Of these 98 experienced a drawdown of more than 4 %, which accounts for about 1 out of 3. Average gain if these were opened as 2nd positions 2.31 %, 72 profitable.

If one were to use this doubling down approach, at what point would you double down? Is it when the trade has gone against you 4%?

I think it is interesting that of 98 trades where they went against you by 4% or more, 72 become profitable not too much later. I suppose this is because the filter picks stocks that are in a sustained uptrend ( ROC(80) > 20 ).

I wonder if another way to mitigate big drawdowns is to scale out. After a trade goes down by x percent, sell off half of the position.

I don't think SF can simulate these second position scenarios, whether doubling down or scaling out, right? Is there other simulation software that can?





Kevin_in_GA
4,552 posts
msg #115672
Ignore Kevin_in_GA
10/2/2013 9:50:39 AM

While it is a very different system, I will occasionally double down on a trade in the Pangolin Z system, but only once and only if the stock has dropped by more than 8% from the entry price. This is always recovered some losses, and occasionally turned a losing trade into a small win. I have not optimized this and am not aware of any software that can do this automatically.

mahkoh
894 posts
msg #115680
Ignore mahkoh
10/2/2013 2:49:05 PM

"In my example from 5/20: Originally 282 trades from 5/18/2012 until 5/18/2013. Of these 98 experienced a drawdown of more than 4 %, which accounts for about 1 out of 3. Average gain if these were opened as 2nd positions 2.31 %, 72 profitable"

It was the second position that was profitable 72 out of 98 times, that does not mean that the whole trade turned out positive. But is does indicate that you have a good chance of at least recouping some loss.

mahkoh
894 posts
msg #115939
Ignore mahkoh
modified
10/15/2013 2:12:21 PM

I have always liked this filter for its sheer simplicity (and its returns, obviously)
However there has also been one thing for me to ponder: The filter assumes that a 2% drop in rate of change has the same impact on a stock, whether it trades for $5 a share or $250 . I don't believe that to be the correct and have been looking for a way to make the ROC requirement variable; the lower the stock's price, the higher the drop needs to be before the filter triggers. I think I have found a way using the square root of the stocks price in a formula:

Fetcher[
S&P 500
set{root,pow(close,0.25)}
set{rocroot,-7/root}
add column rocroot
ROC(7,1) crossed below rocroot

ROC(80,1) above 20
close above MA(200)
add column ROC(7,1)
sort on column 5 ascending

draw ROC(7,1) on plot ROC(80,1)
]



These settings require a stock that trades at $5 to drop about 5 % before triggering while a stock above 150 will only need to drop about 2 %.

tonyctl
7 posts
msg #115950
Ignore tonyctl
10/15/2013 6:41:59 PM

This is a great idea! What should the exit filter look like if going by the square root of ROC?

mahkoh
894 posts
msg #115968
Ignore mahkoh
10/16/2013 1:40:18 PM


Fetcher[
symlist(enter comma separated stocks here)
S&P 500
set{root,pow(close,0.25)}
set{rocrootexit,7/root}
ROC(7,1) above rocrootexit
]



However, keep in mind that I just picked some values that result in what I THINK to be correct settings.

StockFetcher Forums · Filter Exchange · BUYING THE DIP<< 1 ... 4 5 6 7 8 >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus