StockFetcher Forums · Filter Exchange · Telling the Future...<< 1 2 3 4 >>Post Follow-up
heyen
124 posts
msg #43693
Ignore heyen
5/13/2006 3:53:36 PM

Ok, crazy idea, but i was surprised...
c'mon, what would you give to know the future???

Fetcher[
market is nasdaq
set{c1,count(ind(qqqq,day change) 1 day ago above 0,1) + count(day change above 0,1)}
set{c2,count(ind(qqqq,day change) 1 day ago below 0,1) + count(day change below 0,1)}
set{correl,count(max(c1,c2) equal 2,100)}
add column correl
sort by column 5 descending
]



... donations accepted :-)


hscott
34 posts
msg #43696
Ignore hscott
5/13/2006 4:58:17 PM

Changed your filter a little:

set{c2,count(ind(qqqq,day change) 1 day ago below 0,1) + count(day change below 0,1)}
set{correl,count(c2 equal 2,100)}
add column correl
correl less than 6
and average volume(20) greater than 50000
and price greater than 2
sort by column 5 descending

This selects stocks that go up when Nasdaq fall. Backtests pretty well with 1% stop


heyen
124 posts
msg #43697
Ignore heyen
modified
5/13/2006 5:56:33 PM

But isnt MNCS a little non-volatile for day trading?
I'd rather go with

Fetcher[
market is nasdaq
set{c1,count(ind(qqqq,day change) 1 day ago above 0,1) + count(day change above 0,1)}
set{c2,count(ind(qqqq,day change) 1 day ago below 0,1) + count(day change below 0,1)}
set{correl,count(max(c1,c2) equal 2,100)}
add column correl
and average volume(20) greater than 50000
and price greater than 2
and count(day change between -1 and 1,20) below 10
sort by column 5 ascending
]



Interesting: no stock correlates (lag 1) to the QQQQ more than 36% ....




hscott
34 posts
msg #43698
Ignore hscott
5/13/2006 6:50:52 PM

Yes it probably is not volatile enough - but I don't daytrade.
(I didn't realize that SF was that great for daytraders since it uses EOD data)

Heres a mod:

set{c1,count(ind(qqqq,day change) 1 day ago above 0,1)}
set{c11,count(day change above 0,1)}
set{c12,c1+c11}
set{c13,count(c12 equal 2,100)}
set{c2,count(ind(qqqq,day change) 1 day ago below 0,1)}
set{c21,count(day change below 0,1)}
set{c22,c2 + c21}
set{c24,count(c22 equal 2,100)}
set{c23,2*c24}
set{c3,c13-c23}
c3 greater than 7
add column c3
and add column c13
and add column c23
and average volume(20) greater than 50000
and price greater than 2

This weights a fall with Nasdaq higher. Also backtests well


heyen
124 posts
msg #43699
Ignore heyen
5/13/2006 7:10:13 PM

Or have you ever asked yourself, if there are stocks foretelling the market?
Fetcher[
market is nasdaq
avgvol(90) above 25000
set{c1,count(day change 1 day ago above 0,1) + count(ind(qqqq,day change) above 0,1)}
set{c2,count(day change 1 day ago below 0,1) + count(ind(qqqq,day change) below 0,1)}
set{precaution,count(max(c1,c2) equal 2,100)}
precaution above 60
add column precaution
sort by column 5 descending
]



at a "being right rate" between 60% and 70% i'd propably take a democratic vote of the majority for being right.


shelupinin
120 posts
msg #43707
Ignore shelupinin
5/14/2006 4:03:26 PM

heyen, your last filter may show stocks (if precaution is, say, above 60) which has strong correlation with QQQQ, but what is practical advantage to know such stocks? sorry, may be I'm too stupid, but I can't understand... ok, you know some stock wich has high correlation with qqqq, for example AMLN, so How you may predict the future price of AMLN knowing current QQQQ and AMLN???
Alex


hscott
34 posts
msg #43708
Ignore hscott
5/14/2006 4:21:15 PM

If I may add something here.

My mod of heyen's filter was designed as a reaction to the crash last week.
Looking for stocks that go up when Nasdaq goes up but do not fall when Nasdaq falls. So I "add points" for a positive correlation with Nasdaq uptrend and "subtract points" for a correlation with Nasdaq downtrend. Seems to work pretty well. Backtest with 25% stop gives about a 65% ROI but over 90% gains which I don't see that much.

Approach Information
Approach Name: set{c2,count(ind(qqqq,day change) 1 day ago below ...
Test started on 12/30/2005 ended on 05/12/2006, covering 91 days
Filter used:
set{c1,count(ind(qqqq,day change) 1 day ago above 0,1)}
set{c11,count(day change above 0,1)}
set{c12,c1+c11}
set{c13,count(c12 equal 2,100)}
set{c2,count(ind(qqqq,day change) 1 day ago below 0,1)}
set{c21,count(day change below 0,1)}
set{c22,c2 + c21}
set{c24,count(c22 equal 2,100)}
set{c23,1.8*c24}
set{c3,c13-c23}
c3 greater than 12
and Historical Volatility(100,1) less than 40
add column c3
and add column c13
and add column c23


Trade Statistics
There were 190 total stocks entered. Of those, 103 or 54.21% were complete and 87 or 45.79% were open.
Of the 103 completed trades, 95 trades or 92.23%resulted in a net gain.
Your average net change for completed trades was: 6.07%.
The average draw down of your approach was: -3.07%.
The average max profit of your approach was: 7.54%
The Reward/Risk ratio for this approach is: 12.21
Annualized Return on Investment (ROI): 80.50%, the ROI of ^SPX was: 7.97%.

Exit Statistics
Stop Loss was triggered 1 times or 0.97% of the time.
Stop Profit was triggered 46 times or 44.66% of the time.
Trailing Stop Loss was triggered 0 times or 0.00% of the time.
You held for the maximum period of time (150 days) 0 times or 0.00% of the time.
An exit trigger was executed 56 times or 54.37% of the time.

Statistics By Holding Period
 Completed2 day chg5 day chg10 day chg25 day chg40 day chg
Winners:958811011411643
Losers:89478726530
Win/Loss Ratio:11.88:10.94:11.41:11.58:11.78:11.43:1
Net Change:6.07%0.08%1.13%2.18%3.67%2.04%

Statistics By Variable: Match Price
 <20<40<60<80<100<120<140<160<180<200
Completed24:231:522:110:03:04:0-1:0--
2 day chg24:2419:3423:1411:156:33:20:22:0--
5 day chg25:2428:3029:815:116:34:12:01:1--
10 day chg23:2630:2727:1019:67:24:12:02:0--
25 day chg24:2431:2625:1021:37:24:02:02:0--
40 day chg14:410:169:57:43:1-----

Statistics By Variable: Average Volume
 <1000000<2.0M<3.0M<4.0M<5.0M<6.0M<7.0M<8.0M<9.0M<10.0M
Completed82:89:02:0-1:0----1:0
2 day chg77:806:103:30:11:0----1:0
5 day chg96:6710:63:30:11:0----0:1
10 day chg96:6511:54:21:01:0----1:0
25 day chg97:6012:34:21:01:0----1:0
40 day chg32:259:32:1-0:1-----

Statistics By Variable: ADX(14|14)
 <10<20<30<40<50<60<70<80<90<100
Completed3:019:021:324:117:28:12:01:1--
2 day chg1:222:1619:3216:2116:1810:32:01:21:0-
5 day chg1:227:1123:3020:1824:1110:31:13:21:0-
10 day chg1:228:1023:2924:1424:1110:21:12:31:0-
25 day chg2:131:725:2728:1119:129:21:11:4--
40 day chg1:09:210:98:912:72:1-1:2--

Statistics By Variable: Comparative Relative Strength(^IXIC|10)
 <0.96<0.99<1.02<1.05<1.08<1.11<1.14<1.17<1.2<1.23
Completed0:110:020:127:016:515:04:01:11:01:0
2 day chg0:210:620:3123:2319:1911:82:32:10:11:0
5 day chg0:211:937:1628:1817:2112:61:53:00:11:0
10 day chg2:013:638:1431:1516:2211:81:51:10:11:0
25 day chg0:213:835:1633:1022:149:101:41:11:01:0
40 day chg0:14:113:710:89:65:50:10:11:01:0

Statistics By Variable: Historical Volatility(100|1|365)
 <4<8<12<16<20<24<28<32<36<40
Completed-5:04:06:14:112:112:320:120:112:0
2 day chg-5:33:712:147:811:1911:816:1011:1412:11
5 day chg0:16:48:415:118:613:179:1019:814:1118:6
10 day chg0:18:19:217:97:812:1813:619:815:914:10
25 day chg0:15:37:520:69:615:1513:619:613:1015:7
40 day chg0:15:32:33:34:58:53:24:39:35:2



heyen
124 posts
msg #43711
Ignore heyen
5/14/2006 5:46:21 PM

@ shelupinin

Precaution?
Fetcher[
market is nasdaq
avgvol(90) above 25000
set{c1,count(day change 1 day ago above 0,1) + count(ind(qqqq,day change) above 0,1)}
set{c2,count(day change 1 day ago below 0,1) + count(ind(qqqq,day change) below 0,1)}
set{precaution,count(max(c1,c2) equal 2,100)}
precaution above 60
add column precaution
sort by column 5 descending
date offset 5/8/2006
]



You see how our boys dropped in price all together?

Fetcher[
volume equal ind(qqqq,volume)
date offset 5/8/2006
]



And our market index was a little late... to be precise: 1 day late!
Now we know that the qqqq tells the rest of the market where to go...


@ hscott

nice work! my backtests weren't as good.



shelupinin
120 posts
msg #43716
Ignore shelupinin
5/14/2006 7:20:18 PM

ok, heyen, thanks
according to your filter most stocks dropped in price at friday, May 12. So at Monday, May 15, index QQQQ also should drop .... lets see...

Alex


heyen
124 posts
msg #43719
Ignore heyen
modified
5/14/2006 7:46:09 PM

Hi shelupinin,

well, for the statistics sake, thats what I expect.
This could be interesting for day trading as to wether go long or short with the first initial position in pre market or at the opening. Because thats where the most volatility is. I will watch this the next days to see how that works until reversal time.
I have missed the biggest move waiting for the reversal time to pass by more than once.

Thomas


StockFetcher Forums · Filter Exchange · Telling the Future...<< 1 2 3 4 >>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