Herky_love |
Gold User, Member, TeleChart
|
Registered User |
|
|
|
|
Unsure |
|
Wednesday, February 2, 2005 |
Friday, May 24, 2013 9:21:14 PM |
82 [0.03% of all post / 0.01 posts per day] |
|
I'm guessing that coinditions for Eazyscans are evalluated FIFO (first-in First-out) and for each ticker symbol. From this one could percieve an inherent logical AND operator existing between conditions in an Easyscan.
It would seem prudent to arrange conditions in an Easyscan whereby the most likely condition to be false should be evaluated first. In this manner, the engine will progress through the symbol-list most efficiently witout spending CPU time time evaluating conditions.
Given that there are no logical operators available for Easyscan condition evaluation, a short-circuit operator should be avaiable w/in PCF, i.e., AndAlso and OrElse. These type of logical operator terminate evaluation of an expression at the earliest opportune point. The former exits evaulation if the first argument is false, and for the latter if the argument is true.
For example, If 0 > 1 AndAlso (bla bla bla and (bla bla bla / bla bla) * LOG(blablalblabla / sign (blabla bla).... etc.
It never even looks at the sutff after AndAlso. If, however, in the case of 0 < 1 OrElse (bla bla bla and (bla bla bla / bla bla) * LOG(blablalblabla / sign (blabla bla).... etc.
It doesn't evaluate the stuff after the OrElse given that the OR condition has been satisfied.
What I'd lke to see would be a means to arrange coindittions in a scan with the up / down arrows. As it stands adding a new condition - after a scan has been created with multiple conditions - having a high likelyhood of falseness: it will get evaluated last. To fix it one has to re-create the entire scan condition set from scratch.
|
Two things to note here:
-
changing PCF for watchloist column
-
changing PCF for column that is a component of saved columns
In in the first case you must delete the old column using the old PCF and add the column back. In the second case you must load the saved column, delete the old PCF column, add the new PCF column, and save the column set. Then you must load the saved column for the watchlist. You must do that for all the watchlists that utilize the old column or saved column set.
QWEXION: how does PCF dependent indicator work in that regard? There are at least two cases:
-
PCF used as an indicator
-
PCF as indicator that's been saved (accessed in My Saved Indicators)
Does the user have to plot new indicator if the indicator has been changed, i.e., delete the old indicator (either PCF or saved indicator) and then add new indicator and choose the saved indicator or PCF?
That could cause some problems for users if they're not aware such changes don't propagate throughout all their saved stuff that uses that.
FYI for users that aren't aware: When constructing custom indicators they should be saved after plotting said custom PCF as an indicator on a chart in order for them to appear in My Saved Indicators. Any indicator can be saved with custom parameters by right clicking on the indicator displayed at the top of the chart and select 'save indicator'. A custom PCF plotted as an indicator is just that: a custom PCF. ALL parameters for any indicator get saved as one of the My Saved Indicators, e.g., line color, line style, thickness, specific discrete parameter values associated with any arbitrary indicator. A user could establish 5 different custom indicator Bollinger Bands each with a different base period, line style, color, etc
My Saved Indicators is very cool-guy stylin'. I like. I'm mashing the 'like' button on saved indicators.
|
Is that putatively what the 'show value' checkbox is intended to do? It seems to be either broken, i.e., if its putative purpose is to govern whether the Fib % value is displayed or not - if so its definitely broken - or I'm just not understanding how to invoke displaying the value of the Fib % level..
|
I like the Fib Time zones in v12.3 You define a time period as the first seed, i.e., 0 to 1. The first time period after that, is also labeled 1, ie., the first 1 is equidistant in time between 0 the second 1. The time zone sequence progress thereafter according to the Fibonacci progression, i.e., 3, 5, 8, 13, 21, etc. So if perio 0 to 1 was one month, then the Fib Time line labeled 13 would appear 13 months after 0 to 1.
An interesting feature of ALL Fib drawing tools, one can grab the anchor line and drag and drop it. After that the end-points can be reselected and re-anchored at will. That's usefull if placing drawing tools on a chart where one has to be in large time scales initially; the chart will ony go 500 bars back at whatever time scale is being displayed. if one desires to fine tune the anchor points one can drill into the time scale, use the time scroll feature (grab the time scroll button at the bottom of the chart - carefully so as not to click on the tabs down there (that messes everything up LOL) - and slide the entire time scroll bar to the left until the drawing tool comes into view), then click the anchor line and drag & drop it (just slightly), then select the end-point and put it where you want it precisely. You want to try to drag it back to its initial position though (so that the initial endpoint isn't dislodged).
Very cool-guy.
|
That's certalnly a very nice feature.
Currently I have a layout group of several tabs. On one tab I have two watchlists and a chart template. The chart template is linked to the master watchlist via the Blue Group. The master watchlist is a custom watchlist of only market groups. I have created custom watchlists containing all the sub-industries of the major market groups (named by MG number). The secondary watchlist on this layout tab is linked to the red group.
On another tab, the watchlist is linked to the Red Group (as is the chart template on that tab).
After customizing the chart and saving it, I then saved that layout group.
Now when I launch the layout group, the master watchlist of groups shows up sorted by custom column indicators, the chart is linked to the master watchlist active item. I can then open the sub-industry watchlist in the secondary wathclist panel (and it'll be sorted according to custom column indicators). When I select one of the sub-industries in that secondary watchlist, the components of the subindustry appear in the watchlist panel on the other tab, and the chart on the other tab is linked to the active component of the components watchlist on the components tab.
Very nice.
|
Oooh. My bad: I thought #4 was to for the fib% lines. Meh, not so much #4 then. Allow custom fib% lines thickness, dashing and coloring (allow the same for anchor line if you insist). ALSO: allow font size change for fib%
Fib Fan needs something similar. Fib Fan% should be on right side.
|
What he said: Number FOUR!!!! Number FOUR!!! Number FOUR!!!!
This is another excellent example of before bothering people with qwexions, follow the Golden Rule re:ANY software - read the rulebook: first and then peruse the discussion forums (mash that forum search button relentlessly).
Things that are discovered on the way to finding out what you really wanted to know: priceless.
|
It happen: 'change list/scan properties', edit columns, add column
Select 'My Saved Columns': is empty
IF instead click on 'Change Colums' saved columns do appear in 'My Saved Columns'
|
I just discovered this issue also. Decent data re: NYSE composite index can be downloaded in CVS format from:
http://finance.yahoo.com/q/hp?s=%5ENYA+Historical+Prices
WARNING: watch out for the speed-bumps at 12/12/11 & 06/01/12
The data available apparently beings 31 Dec 1965. I took a look around the inter-webs and those two dates are just as speed-bumpy elsewhere (no joy in easy fix for speed-bump data). The thing is: I have some data from TC2000 v12.3 (its just off by some factor, right?). Albeit TC2000 reported NASDAQ volume is off by some 10%, NYSE reported volume is off a whopping +75%; maybe you guys can - I cant - land the Apollo 13 guys in the shuttle with those kind of numbers.
So I fingered out an averaged factor between the TC2000 data compared to the downloaded CVS data for those particular two days - two days either side of the missing data downloaded - and applied that averaged factor to the TC2000 reported value as a subsitute for the missing data in the downloaded CVS format-file.
You guys prolly should beat me with the Statistics Rule-Book - thow out bad data - and by God I prolly need it. But that's my story and I'm stickin' to it (and my spreadsheet charting tool is a happy camper)..
QWEXION: who stole ~4.5 BILLION shares of NYSE shares - eaches - on both of those two days? And what significance is attributable to that?
|
I'm sorry if I've violated proprietary information disclosure policy / rules / law.
Please delete any information that may cauase Worden, Inc. legal troubles.
|
|