What's happening above me?

You will soon be looking at a powerful new way to explore the stock market. The applet above you provides zoomable, fast charts of any of more than 8,000 stocks traded in the United States. Click and drag out a rectangle, and the view zooms in to reflect the new rectangle. This way, you can drill down to see a particular time period. Also, you can chart stocks that are no longer traded. Want to see what happened to 3com? No problem! Just type "coms--0" into the ticker box. That means "The most recently delisted stock formerly known as coms". You can see Delta airlines before, during, and after its bankruptcy. Just type "dal--0", "dalrq", and "dal" respectively, into the ticker box. We have data going back to January 2003.

But that's not all. The applet (when it finishes loading) allows you to try different strategies, and see the modified chart, as your strategy is applied to your stock. You will be able to see charts of IBM only on Mondays, and see how that strategy pans out. Or you can test the "January effect", and see what happens if you buy stocks only in January. Or all months except January. All that is required is for you to type "month != 1" in the term box, and for any given stock that you test, you will not buy during January. This will appear as a fairly normal looking chart of the stock in question, but with a long flat line each year where all the activity in January is suppressed.

Is it there yet?

If you have a Java logo, and progress bar, then everything is likely fine. If the progress bar is moving, that is. If it's stopped, that could be caused by several things. One is an unexplained system error (network problems or the like). Another one, however, is that this is a signed applet, and it needs your approval to keep going (don't we all). All you have to do is indicate your acceptance of the signed applet on the dialog that popped up. Ah, you say, but the wrinkle is that there is no dialog box. That may be because the dialog box got pulled over on its way to your screen. Many browsers come with a popup blocker (look for text like "149 blocked" in your menu bar area). If Google Toolbar or Yahoo Toolbar is installed, it may have its own popup blocker installed. Clicking on the offending text will pull down a menu with options like "enable popups for this site", or similar. Go ahead, there's no porn here. We don't even have any cute pictures of cats hiding in your computer.

With all that done, a reload should get the page to load without further nuisance. If course, none of these issues may have come up. If so, good for you! With respect to the applet security box, "Trust always" is actually more secure than "Trust just this once", because if an evil hacker tries to impersonate this site, their certificate will not be the same as mine, and you will get a new popup that you weren't expecting, which should alert you that the game is afoot.

An error occured. What now?

If you do not have Java installed, you will be unable to run this applet, and unable to use this software. Visit http://java.sun.com, or follow the instructions that the browser provides for downloading and installing the JRE. Any JRE that is version 1.4 or later will work just fine. We do not support the Microsoft VM, because we use software that does not support the Microsoft VM.

If you hava Java installed, and it still does not work, I very much want to hear about it, so I can add good instructions here. Linux users particularly are prone to having almost Byzantine failures, and I'd love to catalog and document the most common of those. If you are a Windows user, and you had to install Java manually, I'd love to hear about it, so I can get a feel for how common that is.

Why don't you use Flash? Or GWT? Or AJAX in general?

We do use Javascript, at least in a vacuous sense. These instructions will disappear when the applet finishes loading :-). However, the deeper answer is that the charts that are being sent back and forth have thousands of data points, per data set, and multiple data sets per chart. Plotting that many points in Javascript would be torture, and would therefore only run on computers in Guantanamo Bay. Flash is faster, but the user interface capabilities are not as rich as I'd like. Also using Java for the user experience technology opens the way to desktop applications, or to interesting hybrids, where as AJAX and Flash are very tied to a browser. Finally, AJAX is very browser dependent, even with GWT.

If you've tried everything above, contact us.

I'm hoping that you're not reading this hoping I'll say something to make it work for you :-).

What is all this for?

You have no doubt noticed three text boxes, and a chart.

The ticker box:

This is where you enter in your favorite ticker. Some choices are presented in a table below, but any of the 8,000 most important stocks in the U.S. should work. Delisted stocks also work, for instance dal--0 for the Delta airlines that existed before it went bankrupt. The rule for delisted stocks is sym--n where sym was the symbol that the delisted stock was known as, and n is which one you want. For instance, if 3 companies all has the symbol xxx and went bankrupt (or changed tickers), then xxx--2 would give you the oldest xxx, xxx--0 would give you the most recent one, and xxx--1 would give you the one in the middle.

Name	Ticker	Notes
S &P index fun	spy
Nasdaq 100 index fund	qqqq
Nasdaq index fund (previous)	qqq--0
3com, inc	coms--0
Yahoo, Inc	yhoo
Merrill Lynch	mer
Valueline, Inc	valu
Evergreen associates	eee	Going down
Apple, Inc	aapl	Going up
Ford Motor Company	f	Cyclical, loser
Exxon Mobile, Inc	xom	good stock to own
Elan Corporation, plc	elan	Very volatile
Dryships, Inc	drys	long term volatility
TVI Corporation	tvin	Can you keep your gains?
Westwood One, Inc	won	You definitely have not won with this one
3M company	mmm	seems cyclical, nice dividend
Tyco International, Ltd	tyc	Ouch in 2007
Bear Stearns	bsc	Ouch in 2007
Brazil etf	ewz	Nice to own

The term box:

The term box allows you to enter conditions under which to hold the stock. These are just expressions that are either true or false at a given time for a stock.

Examples

Lets look at a simple term:

close / close <- 365 > 1.0

This means the closing price is higher than it was a year ago. Lets break it down in the table below:

close <- 365	The closing price 365 days ago
close / close <- 365	The current close divided by the close of 365 days ago
close / close <- 365 > 1.0	The current closing price divided by the closing price of 365 days ago is greater than 1.0

Now we can look at a more complicated one.

(close / (close avg 30)) / ((close / (close avg 30)) avg 100) > 1.0

This term asks "Is the close more above its 30 day average than is typical during the last 100 days? We tear this one apart below:

close avg 30	The 30 day moving average of close
close / (close avg 30)	The ratio of close to its 30 day average
((close / (close avg 30)) avg 100)	The 100 day average of that ratio
(close / (close avg 30) / ((close / (close avg 30)) avg 100)	the 30 day average versus the 100 day average of the 30 day average
(everything avove) > 1.0	Is the above quantity positive

Glossary of fields

field	Explanation
close	The closing price of the stock
open	The opening price of the stock
high	The daily high for the stock
low	The daily low for the stock
volume	The daily volume of the stock, in shares
adjustment	The split adjustment factor for the stock price
pe	The price per earnings
eps	The earnings per share
pctyield	The divident yield for the company, in percent
dividend	The last actual divident, per share
divdate	The date of that last dividend
numshares	The number of shares outstanding
futureeps	The expected eps for this company
day	The day of the month (e.g. 15 if its april 15th)
month	The month (eg 2 for February)
year	The year (i.e. 2008)
The weekday	0 for Monday, etc

Glossary of operators

For the purposes of the table below, A or B may be terms, fields or constants.

Operator	Explanation	Returns
&&	A && B is true if A is true and B is true	1 or 0, 1 meaning true
||	A || B	1 or 0, 1 meaning true
after	A after B is true while A is true, but only after B becomes true	1 or 0, 1 meaning true
before	A before B is true while A is true, but only before B becomes true	1 or 0, 1 meaning true
until	Like A before B, but it remains true even after A is no longer true, until B becomes true	1 or 0, 1 meaning true
then	Like A after B, but persists while B is true even after A becomes false	1 or 0, 1 meaning true
>	A > B is true when A is greater than B.	1 or 0, 1 meaning true
<	A < B is true when A is less than B.	1 or 0, 1 meaning true
>=	A >= B is true when A is greater than or equal to B.	1 or 0, 1 meaning true
<=	A <= B is true when A is less than or equal to B.	1 or 0, 1 meaning true
!=	A != B is true when A is not equal to B	1 or 0, 1 meaning true
==	A == B is true when A is equal to B	1 or 0, 1 meaning true
max	A max days equals the largest value in A within days days	A number
min	A max days equals the smallest value in A within days days	A number
avg	A avg days equals the moving average over the given number of days	A number
rank	A rank days gives the rank within the given number of days	An integer
freq	A freq days gives the number of times A is true over the given number of days	An integer
/	A / B is A divided by B	A number
*	A * B is A times B	A number
**	A ** B is A taken to the Bth power. B is assumed to be an integer	A number
//	A ** B is A taken to the Bth root. B is assumed to be an integer	A number
<-	A <- days is A shifted the given number of days into the past	A number
->	A -> days is A shifted the given number of days into the future	A number

Choices

If you use curly braces, such as

close > {10.0,12.0,15.0}

that represents a choice. The term chosen might be the one with 10.0, or 12.0, or 15.0. Multiple curly brace terms can be used, and can represent a very large number of terms:

({high, close, (close / close <- 5), (close avg 5), (close avg 10)} avg {9,15,21,30,45}) / close > {0.95,0.97,0.99,1.00,1.01,1.03}

represents 150 possibilities.

The count box:

Limits the number of choices plotted on the graph. Keeping this below 10 is a good idea.