Sunday, February 26, 2012

Meta-Trader - Meet Hubert Senters

Welcome Back Meta-Traders.

This past week I had the opportunity to attend the Traders Expo in New York City. I find these events tremendously helpful since I get to see face-to-face some of my favorite market personalities and gain valuable insights as to how they approach their craft. If you have not been to one of these shows, I highly recommend it. You can find the organization at http://www.moneyshow.com and they do 12 shows a year all around the globe. Admission is free and while there is plenty of stuff for sale, you can take as much information and learn as much as you can for free in exchange for your contact information.

I don’t know about you, but I labor in obscurity most of the time. I do my day regular job, and take care of business, both home and family. But underlying all that, there’s a part of me that wants to spend all day, every day with people who make their living in the markets. I try to talk about it with my family, but most of the time, they give me a blank stare and change the subject. These shows give you a chance to hear presentations from those who have made it big and how they did it so you can associate with those people - even if just for one day.


So with this post I’m going to start a new “Meet the Trader” series. In each post, I will cover one trader or personality that I particularly enjoy and what I learned from them.

In this post, we meet Hubert Senters from http://www.tradethemarkets.com. Hubert partners with John Carter and others to run the Web site which is an advisory and trading community focused around Futures, Stocks and Options and Forex to a lesser degree. I got to see a presentation by Mr Senters after knowing him for some time through the free daily trading videos you can get from their website. He also hosted a 2+ hour live trading Webcast last year on the Friday after Thanksgiving when opened his trading room to the public via conference call and live meeting. That said, I’m not a customer of their site nor have I spent any money with them.

Anyway, Hubert is a kindly southern gentleman with a disarming personality and a keen sense of humor. He trades Futures only. At the expo one person kept asking him, why not Stocks or Options? He replied he just doesn’t like them and when pressed, he mentioned the favorable tax treatment we get from Future profits in the United States versus stocks. There’s another better reason that I’ll get to below.

Hubert is a self-made man and trades his own account, full-time from the basement of his house. He describes it as the “Bat Cave”, an 1100 square foot space with high security – visitors have to be “buzzed in” from the outside. As for trading platform, he runs Tradestation on Dell Desktops with an array of flat panel displays. He has a device called a TreadDesk where he can walk in place when trading and move the desk up and down as needed. How’s that for a cool way to incorporate our hunter/gatherer roots with a modern way to make a living?

I mention his trading setup because like me he is a bit of a loner and was clearly not comfortable speaking in front of a large group – at first anyway. He was also uncomfortable in a suit and indicates he spends a lot of time trading in just boxer shorts, and sometime goes days without showering – something I can relate to since I work at home a lot – though I usually just skip a day or 2.

As for trading style Hubert is mainly a day trader, but sometimes position trades as well. Here are 3 specific setups (among many) he uses to pull out profits from the markets on a daily basis:

Scalp against the Opening Gap. Hubert indicates that opening gaps in the futures market will close a high percentage of the time. I saw him do this trade the day after Thanksgiving and he made a few hundred dollars in the first few seconds of the trading day. And he was using reduced lot sizes for the presentation.

Another trade is to go long or short the futures when daily program trading levels are breached as measured by Cash to Futures Premium as found on http://www.indexarb.com. Hubert prefers the Dow or Russell Futures versus the E-Mini S&P which he calls a “sprinter” meaning it runs up and runs back and is much more whippy and volatile. He makes the analogy – why try to land a crop-duster on a moving aircraft carrier when you can land it in an open field?

One last trade is more of a position trade and he calls it the “Goodnight Gold” trade. He says he made more money with this strategy than any other in the past 2 years and it goes as follows. Trade the full-size Comex Gold Contract which follows the price of Gold up and down at $1785 per ounce and can easily move $25 in one day. Take the price at 11:30 PM and subtract from the price earlier in the day at 1:30 PM. If the price is more than $3 up or down (but not too far up or down) take a trade in the direction the move. Use a 6 point stop and a 60 point take profit. If the trade loses, you lose $600. If the trade wins you make $6000. This trade can last several days and he sometimes takes profits on the way to 60 points at 18-20 point intervals. Note that with this type of Risk to Reward ratio, the system will lose far more often than it wins.

Hubert makes a lot of money. He has been profitable pretty much since day one and learned to trade from a guy he ran into by accident when he was working as apprentice to be a large animal vet. When he saw the market data screen, and saw the guy had made $150,000 that day, he was instantly hooked. He became the trader’s apprentice and learned from the guy for 18 months before going out on his own. Hubert’s parents worked in a factory, and he could easily make in one day what they make in a year.

Hubert started out with a $5000 trading stake that he pulled together between his wife and some family loans. He turned that into $20,000 in one day, before giving much of that back and ending the day with $11,000. He grew the initial $5000 into 1.8 Million within about 18 months. He tells a funny story about how making all that money messes with your head and makes you buy stuff, houses, cars etc – but remember to put money aside to pay the tax man!

Hubert acknowledges that only a small percentage of futures traders are successful and that the odds are clearly stacked against you. I think the thing I learned most from Hubert is that you don’t need nearly as much capital as I thought to make a living trading. The extra leverage provided by futures forces strict risk management. Someone once remarked that trading stocks is like driving to California, while trading futures is like taking an airplane!

Check out Hubert and John at http://www.tradethemarkets.com and have a great week.

Saturday, February 25, 2012

Meta-Trader - Pushing to highs

Welcome back, Meta-Traders.

We had yet another good week in equities. Gains came from across the portfolio with new all-time closing highs in CACI, IBM and SXCI. We also had positive action in the broad markets with the Dow Jones Industrial Average pushing a 3-year highs, threatening the 13,000 level, but falling back.

All this pushed my portfolio balances up close to levels not seen since last April. And these levels do not account for some withdrawals I did to fund a college savings account and my Forex accounts, so my wealth continues to grow which is a good thing.

Inflation is starting to become a factor once again with gasoline prices at their highest level ever for this time of year. Geopolitical tensions in the middle east are keeping a bid on oil prices. Unlike past cycles, I'm aware of it much sooner and positioned accordingly this past week adding to a position in Oil Services fund OIH. We are going to get some good trend trades in Oil and Gold in the next few months, so pick a few favorites in that group and watch for trend breakouts.

On the Forex side, we had some positive action with a new equity high in Atinalla Full Portfolio and a sharp recovery in Atinalla #3. Offsetting this was a large loss in Coatl H1 which shed about -5.5% for the week. This was a disappointment since it pushed my forex portfolio into the negative for the year.

I am starting to become concerned that poor performance of the Euro Zone is going to lead to poor performance overall in Forex. This is because my accounts -and many of those on Asirikuy as well - depend on trend continuation action in EUR/USD. Lack of good trending action results in losses for those portfolios as entries from trend following signals fail to turn into meaningful trends. Of course, we expect loosing periods and even losing years. But be aware that all of the systems we are trading here performed well over the past 10 years, and the next 10 years could look very different due to fundamental factors. Think about it and let me know your comments.

I am starting to lose interest in developing trading systems because I think more money can be made by drawing conclusions regarding general conditions and positioning accordingly. You saw some of that in last weeks post - which was in retrospect more of a rant.

Also, there is so much to be learned from successful traders. We will see that tomorrow with the first in a new series called "Meet the Trader", so come back tomorrow for that.

Enjoy your weekend.

Friday, February 17, 2012

Meta-Trader - Wagging the Dog

Welcome back Meta-Traders.

It was another good week in equities, not quite as good as last week, but pretty respectable. While I swore off blow-by-blow recaps of my equity trading, I have a story to tell which led me to some important conclusions regarding trading so read on for more.

Apple Computer continued to fascinate this past week. Recall last week that Apple computer threatened the $500 mark, but closed last week short of the mark at $493. Recall also that I was watching the $497.62 level to re-enter on the long side.

This past week was option expiration and price behavior around option expiration is always interesting. Prices move quickly between strikes, but typically settle around a strike price. My challenge with AAPL was to observe the price action and position accordingly.

Monday morning came along and after a big run-up in the pre-market session, the stock opened the day at $499.53 taking out the old high by over 2 points. The stock ran up to about $502 and started to falter. Gaps are often filled, so I took a bearish position by going long the $520 puts looking for a move back to the low $490s. This position seemed to work out and within a few hours the stock was down to $497 and I was in the money 2+ points on the puts. I entered a stop loss for about 1.5 points below where I bought it to limit losses, but did not take profits. I figured unless I was able to take 3X times the profit as I was willing to lose, I was not trading properly.

That trade did not work out and I got stopped out a few hours later as AAPL once again crossed $500 mark and slowly worked its way higher. Having defended the $500 mark, I figured the next test was to the upside, so I went long the Feb 480 calls for about $22 with the stock at about $501.50. This contract was expiring this Friday Feb 17th, which explains the very low premium. I took a similar position in my 401K Rollover account at a slightly higher price and went about my business. The stock closed Monday at about $502 in quiet action.

On Tuesday, the stock opened at nearly $505, a 3 point gap to the upside. During the day Tuesday, the stock traded as high as $509.50 and closed just below that, not far from the $510 strike. After hours, CEO Tim Cook made some comments and there was some other news, and AAPL was up to about $513 before the after-hours session ended at about 6PM EST. I don’t ever recall seeing so much action in the pre-market and post-market sessions which are usually quiet.

Wednesday rolled around and the stock went as high as $518 in the pre-market, but opened the day at $514. After that, AAPL moved quickly to the $520 area – another important strike and started to settle down. Apple had come much farther than I expected, and much faster. Between 10 and 11 AM, prices pushed higher and soon cleared $520 and within minutes were at $523, $524 then $525! Straight to the moon!

Right around this time, I checked the Option chains for the Options expiring Friday and saw something crazy, see above screen shot. The daily volume in the Options was incredible. It was before noon, and already 37,700 of the 520 calls had transacted, and the open interest in that contract was only about 10,700 contracts! At 100 shares a contract, that represents over 1 million shares of stock at this strike price alone, and this is a stock that only trades about 5 million shares a day on average!

After the fact, I added up all the daily volume for the 14 strikes that would fit on the quote board and came to an incredible 197,000 contracts volume for one day. This represents almost 20 million shares stock for that day alone!

Why is this important? As “Dr J” Jon Najarian described to me one time (I was in the audience at a trader’s expo) when the public buys a call, the market makers turn around and hedge that with long stock within seconds. So with the public buying calls like crazy, the market makers had no choice but to buy stock to hedge their positions to flatten their exposure.

At this point, it became clear to me at this run up was feeding frenzy in APPL stock caused in part by the public and momentum players like myself buying the options, and the market makers buying the stock to hedge their exposure. It was a case where the Options were driving the Underlying and a classic case of the “Tail wagging the dog.”

Since it was about noon on Wednesday, it became clear that we had passed the week’s mid-point, and all of these call contracts would expire by Friday. And I figured that very few of these option holders would take delivery of the stock at $50,000 per contract. That means most of these contracts would have to be sold to close, and that would also cause the market makers to close their hedges. In other words, the frenzy on the upside would soon turn into frenzy on the downside. So I slipped out of my $480 calls at about 43 and took about a 20 point profit within about 2 days!

Shortly after, APPL topped $526 and for a short time, $530 was the next strike target. I almost went log the March calls at this but did not. Within another hour or 2, APPL was back down to $523 and I was feeling okay about being out. At that point, I stepped away and for some lunch and talked with my son who was home from school for the day sick.

I looked a short time later, and the stock was down to about $518, wow, what a sell-off! Within another 20 minutes or so, the stock was about $510, Amazing. It was the classic blow-off top, and I managed to skip out unscathed. I was king for a day, even if it was just a one-lot! After that, I took the rest of the week off, because I figured after that trade, it could only go downhill.

The next day, the stock gapped down to open in the low $490’s then find a bottom at $484 before finding a bottom. So what’s the point of my story?

The point is that trading is complicated. It’s a function of Price, and a function of Time, and the Hour of the Day, and the Day of the week, whether it Options Expiration week and about 1000 other things that cannot be fully quantified. I have been programming computers for 30 years - since punch cards were in use – and I don’t think all this information can be programmed. It is also a matter of instinct and experience, and conventional computer algorithms can’t handle that much information, not in my humble opinion anyway.

Here are my conclusions:

1) Automated Trading has limitations

Sure Hedge Funds can use them for high-frequency trading or arbitrage or one of dozens of other uses. But position trading is just too much information for computers to be relied upon to make market-beating trading decisions.

2) Selectivity is key

Trading is about knowing what to do and when. Computer algorithms can’t distinguish between dead times with no news pending the day before a major news event. Sure, it can be programmed, but is this possible or practical?

3) Know the Material

This is perhaps the most important point. Why am I trading AAPL and not the 7000+ other instruments available?

You may have noticed that currency trading results have been terrible lately. Daniel put out a video last week on the issue of trade chain dependency, but I think the answer is much simpler than that. The reason is that the Euro zone is a complete mess and that EUR/USD has performed poorly as a result. EUR/USD historically has a positive correlation with the S&P 500, but that association has failed recently with EUR/USD underperforming due to its poor performance.

Back in the 1990’s I bought some trading software and did a bunch of research around trading methods, buying new all-time highs, range breakouts etc. What I found is that breakout systems worked well if the stock had a good performance. If the stock had a poor performance, sideways down, etc, I had lousy results. This means that what you choose to trade is as least as (but probably more important than) the system you use to trade the underlying!

Take some time to consider my conclusions and let me know your input.

Enjoy your weekend.

Saturday, February 11, 2012

Meta-Trader - Lifting the Bid

Welcome back Meta-Traders.

Sometimes in financial markets, money just pours from the sky. And all you have to is step outside and take a drink, or stick your hand outside with a glass if you don't want to get wet. It was one of those rare times this week in US stocks.

You know I've been watching Apple computer. After the big gap up after earnings back on Jan 25, the stock peaked at about about $455 then pulled back to the $440s. I picked up some stock when it quietly broke $460 last Friday Feb 3rd. But at $460 a share, 100 shares costs about $46,000 which is no chump change. So I did the next best thing and bought the March 400 calls for about $7000 for less than 1 point premium over its intrinsic value.

During the process, I had a chance to watch the Market Depth screen or what used to be called Nasdaq Level 2 for AAPL as shown in the screenshot above. Its a thing of beauty to watch, bids, on the left, offers on the right, time and sales scrolling on the right side. And with an active stock like Apple, its constantly changing with the ebb and flow of supply and demand. Its an amazing amount of information and fun to watch, for a little while anyway.

Anyway, in the level 2 screen above you can see the Bids stacked up on the left with 5 different market makers at the bid and only 1 at the Ask. In that case I would say the stock was well bid. Also, there is not a lot of supply and with so little stock for sale, offers quickly evaporate and sellers are quick to lift the Ask on whatever they have for sale. Why be in a hurry to sell something that is constantly going higher?

When I pulled the above screen shot on Wednesday the stock was at $475. And since the stock made it all the way to $475, what's the odds is going to $500? Pretty darn good if you ask me.

Anyway, the stock closed on Wednesday at $476.68. It opened on Thursday at $480.95 and closed the day Thursday at $493.17. It blew through the $480's without even a daily close in that 10 point range! We didn't see $500 on Friday as I expected, but we had a high print of $497.6 and I'll be watching that price level next week for a further upside breakout.

What's all the fuss with AAPL, are we chasing tulip bulbs here? The stock has a 60% growth rate with a PE of 10. Cramer says you can pay up to 2X the growth rate. If we go that way we get $6000 a share, clearly a ridiculous number. So what if APPL sold at just half its growth rate? In that case the stock would be $1500 a share or 3x where it is now.

Now clearly at some point the law of large numbers kicks in - APPL can't keep growing earnings at 60% for ever. And with a market cap of 428B, its as a bigger market cap than IBM, Amazon and Google combined. But if you look at it in terms of market share, AAPL is not the market leader in the Smartphone or the PC business. And there are surely millions of people in China who want to buy iPads and iPhones.

Speaking of China, my favorite restaurant stock YUM brands came out with excellent earnings this past week (due to strength of sales in China) and the stock briefly got out to a new all-time high but has since pulled back.

On the Forex side, it was a slow week. After a nasty string of losses, we finally had a winning trade in the Atinalla portfolios. Atipaq Full Portfolio and Sunqu also had some good results while Coatl H1 Live and Amachay Demo went into the negative for the year.

Also, I finally got paid the trading bonus due in my FXDD accounts and you can see those deposits in the Atinalla #1 and Atinalla Custom accounts.

That's all for now, enjoy your weekend and get some rest.

Sunday, February 5, 2012

Meta-Trader - Neural Nets - Part 3

Welcome back Meta-Traders. Here is part 3 of my series on efforts to build an expert advisor in MetaTrader that uses Neural Network technology.

When we last left off, I was building and training networks using FannUtil.exe, an EXE compiled using Microsoft Visual Studio 2008. The next challenge was incorporating it into MetaTrader.

My plan was to output the training data to a file in the format that FANN accepts and call the FannUtil EXE from MT4. Finding the Windows API to launch the program was easy enough, but I soon found out that it will not call Windows console programs. So I recoded FannUtil to a new program called FannWin.exe that did exactly the same thing as FannUtil, except was a Windows 32 binary.

Next up was an issue with file paths. Meta-Trader only allows you to write output to the .\experts\files sub-folder. So I had to resort to the Windows low-level file API’s to output to other locations. Next complication was the fact that everything that a program outputs under “C:\Program Files” gets redirected to “%appdata%\VirtualStore\Program Files” which is a feature in Windows Vista and Windows 7. I finally resorted to sticking everything under c:\windows\temp, not very portable, but at least it was easy to find.

Next step was to start calling FannWinn to train and execute the network. After a few tries at this, I realized that the back-tester was launching FannWin.exe before the prior instance of FannWin.exe had finished. I had multiple instances of FannWin in memory. Once I started thinking of ways for MQL to “block” waiting for FannWin to finish, I knew I was headed in the wrong direction. It was also a reminder that launching a Windows EXE requires many times more computer overhead than calling a Windows DLL function. It might work during expert execution, but would never work inside the backtester.

Why not just write a DLL to do all the Neural Network stuff and call the DLL from MQL4 like Daniel did with Sunqu? I could have gone this direction, but didn’t want to have to re-compile the DLL for every change in network topology, such as number input nodes, etc, etc.

So I went back to the Internet and quickly stumbled upon Fann2MQL here: (http://fann2mql.wordpress.com/http://fann2mql.wordpress.com/) which allows you to call the FANN APIs Functions directly from MQL4. After some trial and error with that, I was up and running and had my network training and executing. Along the way, I used a separate program called FannTool.exe to do some troubleshooting and test the networks created by my expert.

After further testing, I decided to re-work the original problem stated in my first post in the series as follows:

dH1 = (iClose(Symbol(), PERIOD_D1, 1)-iOpen(Symbol(), PERIOD_H1, 1)) / dMorningATR;
dD1 = (iClose(Symbol(), PERIOD_D1, 1)-iClose(Symbol(), PERIOD_D1, 2)) / dMorningATR;
dD2 = (iClose(Symbol(), PERIOD_D1, 2)-iClose(Symbol(), PERIOD_D1, 3)) / dMorningATR;
dD3 = (iClose(Symbol(), PERIOD_D1, 3)-iClose(Symbol(), PERIOD_D1, 4)) / dMorningATR;
dW1 = (iClose(Symbol(), PERIOD_W1, 1)-iClose(Symbol(), PERIOD_W1, 2)) / dMorningATR;

Instead of feeding it the last 3 hourly bar changes before the days open, feed it just the difference between yesterday’s close and the morning open. This should be a more indicative value of potential price action for the following day. To put it in words, the inputs are:

  • Change between yesterday’s close and morning open as % of ATR
  • Change between 1 day ago close and 2 day ago close as % of ATR
  • Change between 2 day ago close and 3 day ago close as % of ATR
  • Change between 3 day ago close and 4 day ago close as % of ATR
  • Change between prior week’s close and 2 weeks prior as % of ATR



Network output (and what we are trying to predict) is the daily change between the day open at 8:00 GMT and the day close at 21:00 GMT and will range generally between -1 and +1. Note that in all cases we are using the ATR calculated in the morning at the starting hour since it will likely be different at the end of the day.

Before we get to the results, there are further parameters to consider such as:

- How many days input should I train the network with?

- What is the optimal stop loss value as a % of the ATR?

- What is the optimal threshold for entering new positions in terms of the network prediction?

Let’s take these one at a time.

As for lookback period, I tested 10-years, 90 days, 60 days and 30 days. What I found was that the shortest lookback time period produced the network with the lowest error rate and the most profitable results. The results were not great, but we will get to that later. This was a bit counter-intuitive, because you would think that more data would produce a better results, but it was the opposite and it brings up the perspective that the network isn't really learning actual associations in the data. Its just learning the tendency in the data over the test period and it's predictions change over time based on an underlying trend or bias in the data.

As for stop loss, I tried 0.5 ATR, 1 ATR and 2 ATR. I decided to settle on 1 ATR since 0.5 seemed to get hit too often, and 2 ATR seemed to big a loss to accept.

As for Threshold to take positions, I added 2 properties to the EA, Long Threshold and Short Threshold. I take long positions of the prediction is greater than the Log Threshold and close out longs and go short if the prediction is less than the Short Threshold.

As for values, for the threshold, I tried as low as 0.1 and as high as 0.5. I also coded an option to calculate the 30-day average of the ATR change between 8:00 and 21:00 and take positions if the prediction exceeds the Average Change. I haven't go the full results of those tests yet.

What about performance?

The table on the left shows a 10-year year-by year test using a Threshold of 0.1 for longs and -0.l for shorts. There were some good years, but also some disastrous ones. In all cases the Drawdown exceeds the profits, and given the size of the profits versus the draw downs, it doesn't seem like the network has an edge in predicting future prices.

I also did 10-year back-tests of the system on USD/CAD and USD/CHF and found the results to be about break-even with draw down as high as 50%. So the fact that showed any profit at all on EUR/USD is because that pair tends to have very pronounced trends.

In conclusion, we do not appear to have a profitable system on our hands, and much more work needs to be done.

Check back later and have a great week.

Friday, February 3, 2012

Meta-Trader - All Time Highs

Welcome back, Meta-Traders.

The markets favor those who pay attention. I pay attention to new all-time highs, particularly when they occur in combination with record revenues and record earnings. Nothing gets my juices flowing like when a stock breaks out to a new, all-time high. There’s nothing more bullish that can happen to a stock. It signals there are no sellers left and the price can scamper ever higher, “like a scalded chimp” as Jeff Macke likes to say.

I bring this up because we had new, all-time highs in a few of my stocks this past week such as Apple, HMS Systems, IBM and Yum Brands. Equities are straining to rally higher and investor sentiment is the perfect contrary indicator, with investor interest and volume at lows and exposure to bonds at highs. A free and easy US Federal Reserve, and record low interest rates are the perfect fuel for a bull market in stocks.

Friday’s jobs number ignited another rally with the number of unemployed reaching the lowest level since January of 2009. The broader indices showed strength across the board with the Dow moving out to a 13th month high. Emerging market ETFs EEM and EFA had 5 straight up days showing substantial strength in the Euro zone and worldwide. At this point, the all-time highs set back in 2007 seem within striking distance setting the stage for a move to new all-time highs in the major indices.

On the single-stock side, the upcoming IPO of Facebook has ignited enthusiasm among investors. It’s easy to look at Facebook and dismiss it as a fad. But in his letter to investors included in the S1 filing, Mark Zuckerberg draws a comparison between Facebook and the invention of the printing press. Facebook is a medium that is so transformative, it creates new industries and economies based on an interaction model that did not exist before. And with 800 million users, it’s a huge business. And when that stock first trades, 5 billion dollars in wealth will be created out of thin air. Is this a great country or what?

Forex was ugly this past week with losses in all systems except for COATL H1 and Sunqu Live. At this point, we have now given back all of the gains made in the first week of January. This is well within expectations, so no reason to change our system or strategies. It’s almost like you need to put on a different pair of glasses (or goggles) where you don’t consider short term performance. It took me a while to get it, but I definitely get it now.

One thing about equities is that they always go on sale. This typically corresponds to a geopolitical shock, usually occurring the in the middle east and taking a hit to equities. That will constitute another buying opportunity which is why I always keep about 20% in cash.

That’s all for now, check back tomorrow for part 3 in my series on Neural Networks and enjoy your weekend.

Thursday, February 2, 2012

Meta-Trader - Facebook S-1

Welcome back Meta-Traders.

I'm not normally a fan of buying into IPOs but this one deserves some notice. Yesterday, Facebook filed its S-1 form with the SEC declaring its intentions to go public. The deal is going to be the largest Internet IPO ever and the probably one of the top 5 IPO's of all time. The shares are scheduled to go public some time in the late spring under the symbol FB and will trade on the NYSE.

Facebook is a mega-trend it itself and you could argue is in economy onto itself, in the same category as Amazon or E-Bay. Except that Facebook is bigger than all of them in terms of number of users, page views, etc etc.

Go over to SEC EDGAR database and search for Facebook.

Check back on the weekend for my next installment in my Neural Network series.