March 2008 Archives

links for 2008-03-30

| 3 Comments

links for 2008-03-29

| 1 Comment

Its Sure Looking Like A Bottom, Maybe

| 2 Comments
SP500 Volatility Chart 032808I was checking my volatility chart this morning and noticed a massive volatility spike on 3/18, two days after my timing model issued a BUY signal. The BUY signal flashed on 3/14, I moved money around on 3/17 and the volatility spike happened the following day. This is sure beginning to look like a bottom, something that I posted about in my "Has The Market Bottomed?" post earlier this week. Still though, a lot depends on how the subprime mess continues to devolve and how the general market handles it, but the sentiment seems to be firming and I'm seeing a lot of contrarian indicators pointing to higher markets in the future. This could, of course, be a big head fake and we could easily go down from here. However I plan to rotate money out of my bond holdings over the next few weeks and slowly move money back in, assuming we don't break the recent S&P500 low of 1276! Disclosure: Still 40% in cash and bonds.

links for 2008-03-28

| No Comments

Neural Market Trends Top 10 Posts

| No Comments
I really like Wordpress's popular post plugin. Without it I couldn't showcase my updated Top 10 posts!  Without further ado, here they are:
  1. Calculating Historical Volatility 
  2. Building an AI financial market model - Lesson I
  3. Build an ETF Trend System in Excel
  4. Building an AI financial market model - Lesson III
  5. Building an AI financial market model - Lesson II
  6. Understanding Fuzzy Trend Following in Excel
  7. Building an AI financial market model - Lesson V (updated!)
  8. Building an AI financial market model - Lesson IV
  9. Using Genetic and Evolutionary Algorithms to Build a Trading Model
  10. Monte Carlo Simulations For S&P500 Volatility Timing Model
If you scan the list real close you'll noticed that list number 7 is actually updated!!! I updated Lesson V with new Rapidminer compatible templates and model files! Go check it out! If you're a new reader, why don't you consider signing up for my RSS feed? You can do it in your feedreader or get a daily email with my posts!

Exxon Mobil Gives A Buy Signal

| 4 Comments
Exxon Mobil Corporation (XOM) has been thrashing around for the past few weeks on some high volatility. One moment the price of oil is at $90, the next its at $111, and now its around $100. Although XOM is mostly a refiner, the price of oil affects its profit margins and in turn the expectations of investors/traders. SNM-XOM-032508Loading up my Exxon Mobil neural net model this morning I find that its giving us a BUY signal for this morning's open. This neural net model has been pretty profitable over the last 60 bars, its up over 11% (not including commissions/fees). I've been using Stock NeuroMaster since the beginning of the year and have been very happy with it.  I owe my readers a tips and tricks post on how I use it generate these BUY and SELL signals.  If I have time I'll write up the post over lunch!

GE Finally Moves

| No Comments
GE-032408I've been riding General Electric (GE) up and down for the past year and it looks like it might finally be heading back up. My GE neural net model gave a solid BUY signal on 3/17/08 for the following open and a HOLD signal at yesterday's close. This will work out nicely if a market bottom was put in place on 3/17. If. Now I have to check my EME, C, XOM, and NE models!

links for 2008-03-24

| 3 Comments

Has The Market Bottomed?

| 2 Comments
SP500 Volatility Indicator 032008I'm not sure yet but its beginning to look that way for now. On 3/17/08 my market timing model indicators registered the highest reading since I started running it. The next three days the indicator collapsed as quickly as it shot up. My twitter followers would've known that I started buying funds in my 401k accounts that day and I'm sitting on small gains right now. I remain extremely cautious and still hold about 40% cash and bonds but if the low on 3/17 holds (1276) then we have a good chance to see the markets recover.

Gold Implodes For Now!

| No Comments
I know this is so last week but you have to admire how quickly Gold broke down. The chart reveals that it broke through its 50 DMA within in two trading sessions and could even head lower from here. A lot of people think the top in Gold has happened and I probably agree with them for now.

Gold-032108

The subprime mess isn't over yet, I'm waiting for the other "shoe" to drop in the banking industry. In other words I'm expecting one more bank to fail or require emergency Fed funding to keep operating. Which bank? I'm guessing here but I think it might be Merrill Lynch. This is just a hunch and I have no knowledge of its current financial status. Any more volatility related to subprime or banks failing and you'll see Gold over $1000/oz again!

links for 2008-03-21

| 2 Comments

links for 2008-03-20

| 5 Comments

links for 2008-03-19

| 2 Comments

links for 2008-03-18

| 4 Comments

links for 2008-03-17

| 2 Comments

Bring Back The RapidMiner Videos?

| 6 Comments
I've been getting tons of emails from readers asking me where all my RapidMiner video tutorials are. As you may or may not know, I started making RapidMiner Video Tutorials last year but had to stop because of my illness over the Winter. In between that time, my trial subscription ran out and I decided not to fork up the $299 to pay for Camstasia Studio 5 because of financial issues related with my illness. However, if my readers feel generous and willing to make donations to the Camstasia Studio 5 fund, I will be able to get a copy of the software and produce my new tutorials in video format. I can make these videos a lot faster than I can write up the tutorial posts, believe me!
I've mentioned this before but there are some compatibility issues between YALE 3.4 and RapidMiner 4.0. You can't load YALE 3.4 files into RapidMiner 4.0 directly and have it work correctly. As such, the downloads that are a part of my tutorials won't work in RapidMiner unless you build them from scratch in RapidMiner. My plan is to upload revised RapidMiner tutorial versions over the next few weeks, time permitting. I'm currently balance a massive work load and my time is gotten limited and I'm way behind on my new tutorials. Maybe if I have time at lunch today I'll make some edits to the tutorials...

links for 2008-03-15

| 2 Comments

links for 2008-03-14

| 2 Comments

links for 2008-03-13

| 2 Comments

links for 2008-03-12

| 1 Comment

links for 2008-03-11

| No Comments

links for 2008-03-10

| 2 Comments

VIX Buy/Sell Signals For Option Traders

| 9 Comments
Ah, the almighty ^VIX, everyone's favorite indicator and the one index that a lot of option traders watch with bated breath.  What if I were to tell you that you can use Stock Neuromaster to generate BUY/SELL signals for the ^VIX?  Would you be interested? SNM-VIX-030808I'm sure that option traders and writers would be keenly interested if a^VIX neural net model generated a SELL signal for a new short position because it would mean volatility is expected to drop.  If it generated a BUY signal for a long position, then the option trader would expect volatility to increase! Don't believe me?  Here's the chart! Another way to use this when trading options is to build a neural net model for a particular stock and then buy/sell call and put options against it!

links for 2008-03-09

| No Comments

links for 2008-03-08

| No Comments

Creating Buy/Sell Signals for Gold Futures

| 4 Comments
I wrote in my Stock Signal Reviews - Boeing and The S&P500 post yesterday that you can use Stock Neuromaster to create buy and sell signals for Futures data as well. If you can download your quote data into a CSV format (or at least transform it to CSV), then you can import that data into the program, build a neural net model, and generate buy and sell signals. The only downside is that you have update the model manually by downloading the next day's, or week's, data in CSV format.

snm-gold-030508.jpg

Here's a snapshot of the Gold Continous Futures Contract model I built using one year of data through the close of 3/5/08. You can clearly see that the Gold model closed out the short position for 3/6/08's open. Probably a wise move since Gold might move higher over the next few days.

snm-gold-030608.jpg

Next is a snapshot of the Gold model updated with 3/6/08 data loaded in from CSV formated data. You can see that it generated a BUY signal for the market open on 3/7/08. Unfortunately Gold closed down at $973.20 on 3/7/08 and you'd be sitting on small loss if you went long but the model issued a HOLD signal for Monday's open so we'll see what happens. You can even load in Currency Futures data for all of us Forex traders out there! In fact, I just did that for AUDUSD currency data!
Tip: Read the help instructions on how to format the CSV data (especially for the date format) before you load it into Stock Neuromaster. If you don't formate the CSV data correctly, it won't load and you'll stare a blank screen wondering what happened.

links for 2008-03-07

| 2 Comments

Stock Signal Reviews - Boeing and The S&P500

| 5 Comments
This week I featured the neural net models of Boeing Company (BA) and the S&P500 Index and indicated if each model had issued a BUY or SELL signal recently. SNM-BA-030608For my BA post, my neural net model issued a SELL signal on 2/26/08 (for 2/27 open) to establish a short position. It then issued a BUY signal to close the short position on 3/5/08 (for 3/6 open). This was another successful trade using the Stock Neuromaster system and it would've netted you a gross profit of +4.6% in 6 trading days! SNM-SP500-030608I also gave my readers an S&P500 Trend snapshot from my longer term S&P500 neural net model. The model issued a SELL signal on 2/28/08 (for 2/29 open) to establish a short position. It then issued a BUY signal to close the short position on 3/5/08 (for 3/6 open) . It's too bad the model didn't anticipate the sell off yesterday and then issue the close position signal last night but its prediction of the short trend term was correct! This small trade would've netted you a gross profit of +2.3%! Once again the system proves to be a money maker and the beauty of it is, you can pick what stocks you want to follow! One bonus to the Stock Neuromaster system is that you can download OHLC data for any asset, in CSV format, load it into the system and build a neural net model with BUY and SELL signals! Are you interested in modeling Futures like Gold, Silver, Aussie Dollar, or Wheat? You can do that with this little program!

links for 2008-03-06

| 4 Comments
I was inspired to write this post after I read Foquant's (formerly CPP Trader) post on Inductive versus Deductive Algorithms. He hits the nail on the head when he ends his article with:
Personally, I have used deductive reasoning to develop frameworks for money management methods, and then tested data in a similar, inductive method, to create the details.
There's a reason why I like Foquant's blog, he an I think the same way when we approach model building but he just likes to use fancy terms! :) Because I've been in the corporate world too long, I prefer to use the terms "top down" and "bottoms up," when building a financial/neural net model. However you can easily replace those terms with "deductive reasoning" and "inductive reasoning" respectively. The bottoms up approach is where you have oodles of data and you spend time cluster mining for relationships or for statistically significant patterns. Over time you start building a model that will lead to your output variable. This method is very rigorous and time consuming method but the final model should be very robust. The other approach, top down, is where you a lot of time observing your output variable (i.e. stock, currency, index) behaving in the market environment and try to figure out what makes it work. Once you think you have an idea on what makes your output variable function, you gather the appropriate input variables and then statistically try to prove their relevance. Of course if you find that your inputs aren't as robust as you like them to be, you'll have to spend additional time looking for the right ones. Both approaches have their strengths and weaknesses and figuring out what approach to use to build a model really depends on the individual. While one method tends to be more trial and error (top down), the other tends to be more hypothesis testing (bottom up). Personally I start with the top down approach to build a model and then to check it using a bottoms up approach, just like Foquant. This is perhaps the most time consuming way of building a financial model but its led to great success for me and I continue to use it to this day!

links for 2008-03-05

| No Comments

My Trading and Investment Strategy

| 1 Comment
Bull.jpgYesterday I got a great email from Digital Dude who asked me why I didn't use a stop when I went long Citigroup @ $42/share and does buying more, now, make it right? I responded by saying that my strategy for long term is different than my short term one. Soon after I posted my reply I realized that my answer wasn't very clear to Digital Dude and my readers. Simply put, my strategy in my 401k and IRA accounts is vastly different than my trading accounts, both for stocks and Forex. For my retirement accounts I like to market time the markets (buy in panics, take profits during euphoria), establish long term positions (5 years or more) in specific stocks like XOM, GE, and yes C! I choose these stocks on fundamentals and sentiment mostly and because I believe they're going to be winners 5 years from now.  Just like on my mutual funds, I don't use stops when entering these positions because I expect to be in them for a long time.  I'm always net long in my 401k and IRA accounts. Sometimes this strategy backfires on me, like it did on Citigroup (C). So why buy more? My long term outlook for C still stands, I think its a great company despite its bad decision with its subprime investments. Now that the sentiment is even worse, rumors of another $18 Billion write down are floating around and fundamentally its become more attractive, I'm looking to add a few more shares to my 401k account. My exit strategy for these stocks will probably be bequeathing them to my grandchildren upon my death. My trading strategy is entirely different. I use my neural nets to determine Forex trends or potential breakout stocks, I swing or position trade them, and I use stops! I typically place these stops using my Monte Carlo simulations. This works extremely well in the Forex markets I found out. My exit strategy on these are within a few days and occasionally weeks. question.JPGJust a word of caution, my trading and investment strategies work for me but they may not be right for you. I've learned the hard way over my trading and investing life what works for me and what doesn't. You'll have to find your own strategies.

links for 2008-03-04

| No Comments

S&P500 Trend Snapshot

| No Comments
SNM-SP500-030308This is just a quick snapshot to see where the S&P500 is going. My long term S&P500 neural net model initiated a SELL signal on 2/28/08 (for 2/29 open) and maintains the short position for today's open. I guess my Monte Carlo simulation, with its bias to the downside, is being proven correct. I just wonder if my initial downside target of 1291 will be hit.

Forex Trend Signals

| No Comments
currency.thumbnail.jpgIts been a while since I posted my Forex Trend Signals and the reasons why are numerous: I've been sick, swamped at work, and taking a break from trading Forex. The break in trading Forex was out of necessity. Although my trend signals were correct, the markets intraday were extremely volatile and stopped me out several times in a row. So before I decimate my trading account, I took some time off to think about my strategies before I start trading again. The signals for this week shouldn't come as a surprise to people, they are very obvious if you've been watching the market.
  • AUDUSD - UP
  • GBPUSD - UP
  • EURUSD - UP
  • OIL - UP
  • GOLD - UP
Note, these trend signals are based on proprietary neural net models I've created and they get updated weekly. I use them to guage if I should enter long or short swing positions on intraday or intraweek pull backs. These strategies have worked wonders in the past but lately they've caused me to be stopped out during intraday volatility. What I really need to build this year is a better Forex volatility model to help me predict the direction of volatility and I'll probably do that after I post my new tutorials and have the first RapidMiner Meetup in NYC.

links for 2008-03-03

| No Comments

The Boeing Company (BA) was flying high for a while but recently has been dragged down through this market correction. BA is generally a popular stock and its even been showcased on Wallstrip last week, right before they lost a $40 billion dollar contract to their rival Northrop Grumman. This is a slap in the face for Boeing because Northrop will be using the Airbus’s (their main rival) plane design to replace the Air Force’s aging in-flight fueling planes.

Fundamental Analysis

S&P currently rates BA as a 4 star company and was downgraded from 5 stars at the end of last year. BA pays a $0.40 dividend, its gross margin is 16.1%, net profit is 6.1%, and its PE ratio of 15.8. It has an ROA of 7.3%, an astounding 73% ROE, and a ROI of 16.4%. BA’s earnings grew over 83% for the past year with last year’s earnings being $5.26. Analysts expect BA’s earnings to grow to $5.98 in 2008 and $7.15 in 2009! So far BA looks pretty good and with its recent price pull back, but is it a BUY?

Technical Analysis

The technical picture for BA is so-so on the weekly charts. After peaking around $105, BA has pulled back to the $80 level and below its 50 WMA. It’s looking short term Bearish to me but long term it’s still Bullish for one simple fact, in my mind, it’s above the 200 WMA. But is it a Buy?

BA-022908

Neural Net Signal

SNM-BA-022908My BA neural net model issued a SELL signal on 2/26/08 to open a short position and it remains short going into today’s open. According to the neural net, it’s not a BUY but a SELL!

 

Price Targets

My Monte Carlo simulation indicates a downward bias for BA at this moment. That’s probably why the BA neural net model is short! The short term downside targets are: $78, then $71, and then $64. The most likely objective is the $71 level. Short term upside targets are $85 and $92.

Bottom line

I like BA for my 401k as a long term play but I’d wait to see if BA keeps selling off as indicated by my neural net model and my Monte Carlo sim. If it does and reaches the price objective of $71, I’d look to BUY for my 401k but in the meantime, do nothing.

Disclosure: No positions

Modern Day Ghost Towns?

| 2 Comments
ghost townI roamed around a lot when I lived in New Mexico and I came across several ghost towns. I loved finding them out in the middle of nowhere and poking around in them. If I had more time I would probably researched the reason why they were abandoned but we didn't stay too long at them. Usually they were on private property and had big "KEEP OUT" signs and we had to always watched out for a Rancher with a big rifle. Still though, the sign and the lure of ghost town adventure didn't deter my friend Kevin or me! I mean, how could you not be deterred? I was chatting with the Chairman yesterday when he sent me this link from Bloomberg about the rise in house inventories and how they're at the same level since 1973.
In January 1973, the number of finished new homes for sale was 97,000, when the U.S. population was about 212 million, according to the U.S. Census Bureau. In December 2007, 197,000 completed homes were on the market and in January 2008 there were 195,000. The current population is 303.5 million.
It seems that back in 1973, the % inventory to the US population was 0.05% and today its 0.06%. I really don't know if these numbers are a cause for all the calamity the Bloomberg article alludes too but what it does mean is that home prices probably have further to fall. It's just a matter of time till we work off the inventory and hit bottom and I think we're almost there.

links for 2008-03-01

| No Comments

S&P500: Catching Falling Knives

| No Comments
bear.jpgI make it a habit of "selectively" catching falling knives and yesterday was definitely such a day. I emailed my members yesterday around lunch time yesterday to tell them that my Market Timing model issued a strong BUY signal. Interestingly enough it was at the 1335 level which is right around the 1337 support level I modeled using my Monte Carlo sim. Could the market go lower from here? It sure could and I expect it to kiss the 1291 level,just like it nearly touched the 1383 level earlier this week. So why BUY now? Why not wait till the knife hits the floor? Well its true that this type of investing is counter intuitive. No one can every pick the bottom exactly and I'd rather buy quality mutual funds and stocks that are being sold in panic then waiting for a chart bottom to form. It seems nutty and insane but it works for me.

SPX-022508

I reiterate my downside targets from here: 1291, then 1245, then 1191. You don't want to know the levels below that because they are downright scary!!!

About this Archive

This page is an archive of entries from March 2008 listed from newest to oldest.

February 2008 is the previous archive.

April 2008 is the next archive.

Find recent content on the main index or look in the archives to find all content.