Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
~~~COMPX 12/13/2002~~~~~
Previous close-1,399.55
1375 Phil
1389 timhyma
1398 WTMHouston
1409 AKvetch
1419 Churak
1429 MM
1449 NoMoDo
~~~COMPX 12/13/2002~~~~~
Previous close-1,399.55
1375 Phil
1389 timhyma
1409 AKvetch
1428 MM
1449 NoMoDo
You might enjoy playing with this theory for converting indicator predictions into pricebar high and low predictions. It measures historical differentials from the actual indicator rather than from the predicted value of the indicator.
When the slope of the indicator is positive measure the excursion from the indicator to the pricebar's high. Do the same for the low.
When the slope of the indicator is ‘negative’ measure the excursion from the indicator to the pricebar's high. Do the same for the low.
Store the results separately based on direction of the indicator so that one set can be used for uptrends and the other for downtrends.
Determine the Maximum Favorable Excursion over an optimal number of bars. I used MFE(4). Extrapolating MFE from the ‘predicted’ value of the indicator gives us the maximum predicted high. I also predicted a ‘minimum’ high.
Using the same process I predicted both a maximum and minimum value for the Low.
The result is 2 predictions for the Low and 2 predictions for the High. I lost interest and filed it away when I ran into problems due to my lack of programming skills. I haven’t looked back at my notes but I think about it once in awhile. It makes more sense to me now to apply indicators to the H and L and measure the excursion from there rather than from an indicator applied to the Close. Maybe not, theres value in the Close that's not in the extremes.
What news release
all I see is price <g>
Looking at the pricebar pattern on the $NDX, QQQ and NQ's it seems yesterday was an outside day which followed an inside day.
Actually it's not too late because I'd just want to backtest it rather than use it to base any decision on.
The difficult part has always been knowing which strategy to use at any given time. If you come up with a way that determines a sideways market exists then it's possible to adjust the trending triggers a little wider to avoid getting whipsawed. However that carries with it the disadvantage of increasing the dollar amount of each whip so it's not as trivial to implement as it seems.
That being said, I still like your idea because it's thought provoking. Thanks.
test
~~~COMPX 12/12/2002~~~~~
Previous close-1,396.60
1398 WTMHouston
1409 shao
1414 AKvetch
1422 timhyma
1426 MM
it reminded me a lot of the way AIM works,(excluding AIM's buy/sell action).
So it's similar to the frame work that AIM works within ... but AIM uses a counter trend buy sell strategy within that frame work? Buying dips and selling rallies automatically ...as long as the trend identified by the frame continues higher?
Is the frame fixed in place for a period of time or does it follow the slope of a moving avg and update with every change in the average?
NOTHING works from Yahoo websites.
LMAO!
Richard Dennis and Russell Sands are turtle dove lovers too. lol
Cool chart! Thanks. I just bookmarked all the AIM threads and have plenty of reading to do. I can see quite a few similarities with Darvas. Thanks again, MM
Why do you call it a Turtle Down Trend Line?
Don, I’m not familiar with the AIM formula or variants. Could you give a brief description? It’s my guess Darvas uses actual values while AIM uses stdDev around an optimal avg. One aspect of Zeev boxes I struggle with is outrageous bids. Limit buy orders a good distance below price without being so far away that they break support. That sounds like my understanding of what AIM does too. I think I need to take another look at AIM. I’m particularly interested in how it could build a box on top of a box. Or what causes box values to change. I guess that’s the buy sell safe values?
Regards, MM
The Crazy Train. There's a bug in SI's search engine that proves it. Do a subject search for crazy or train and it pulls up TGTBATU.
Subject search for crazy:
http://www.siliconinvestor.com/stocktalk/searchresults.gsp?s=crazy&domain=titles
Subject search for train:
http://www.siliconinvestor.com/stocktalk/searchresults.gsp?s=train&domain=titles
I prefer TS. I think eSignal needs to increase the intraday data beyond 60 days and add an optimization engine.
10/22/02 is another inside bar.
I didn't analyse the intraday data but the daily shows the bar following it initially had a false breakout. I'm interested in determining at what point of retracing an inside day I could anticipate it would breakout the opposite end.
Sept. 4 wasn't a true inside day but it has the range contraction quality. I'm not sure how to define the variables in a way that it would work with that though.
Take a look at 10/9/02. It's a true inside day followed by a day that initially breaks downward. LMAO!
A range contraction sometimes results in a false initial breakout that leads to a true breakout that carries with it extra momentum because of all the passengers who got on the wrong train. It's a special case study of inside bars. I think it would be preferable to anticipate it rather than try to get on once it's moving with force.
~~~COMPX 12/11/2002~~~~~
Previous Close- 1,390.76
1398 WTMHouston
1406 GGraessle
1411 AKvetch
1419 MM
I am leaning toward thinking that an inside day, suggest another inside day.
Suppose an inside bar breakout indicates direction to a lot of folks who jump on the train. Then the next bar falls back into the inside bar. I'd be interested in knowing which point of retracing the inside bar (percent of inside bar) the breakout entries start to be stopped out with such momentum that it's likely to breakout in the opposite direction. A stop entry that anticipates the 2nd breakout...
Excellent. Your notes are the first attempt I’ve seen on iHub to document a consistent approach. Originally I mentioned Oddball on the MDA thread during a discussion of absolute values. Quite a few people feel absolutes don’t exist in TA.
The discussion was moved here by Chun Li because Oddballs comparison of today’s vol to yesterday’s vol mimics an aspect of Zeev boxes known as specific volume. Specific volume is the 3rd dimension of Zeev boxes.
Oddball’s weakness is it doesn’t use stops or consider price at all. It needs a price movement filter like Darvas. And Darvas needs something non colinear with price like volume.
When the AC compressor cycles on and off it means the freon is full of contaminants. When you get it serviced they have to hook a vacuum pump to it to suck out the old freon before they add new freon. That should do it otherwise the dryer is bad and has to be replaced. I've had the same problem myself quite a few times.
Me too!
Wag higher next time and you won't have that problem.
They're both right.
you think that's a good wag?
~~~COMPX 12/10/2002~~~~~
Previous Close-1367.14
1377 MM
1380 Churak
1388 timhyma
1392 GGraessle
1395 AKvetch
1398 WTMHouston
1402 Albert
1418 shao
~~~COMPX 12/10/2002~~~~~
Previous Close-1367.14
1380 Churak
1388 timhyma
1392 GGraessle
1395 AKvetch
1398 WTMHouston
1418 shao
1422 MM
~~~COMPX 12/10/2002~~~~~
Previous Close-1367.14
1380 Churak
1388 timhyma
1392 GGraessle
1398 WTMHouston
1402 MM
Nice work! Thanks.
~~~COMPX 12/09/2002~~~~~
Previous Close- 1,422.44
1398 WTMHouston
1431 shao
1441 timhyma
1451 MM
Congratulations Shao!
Generically I prefer swing extreme when I'm not describing anything in particular. I used to use pivot points but think it causes confusion with that other strategy that's unrelated to swing extremes. (H+L+C)/3 might be the basis for it and then there's countless derivatives. support1, support2, resistance1, resistance2.
If it holds above c[1] it could be a thrust exhaustion, more commonly called a key reversal. Confirmation would be Monday exceeding today's high.
That's interesting and yes I agree your math converts it to percentages. If you remember where you got the idea I'd like to look into it further.
Nasdaq: $ADVQ minus $DECLQ = $ADDQ.
Nyse: $ADV minus $DECL = $ADD.