The Bulls & Bears of Starting a Stock Business

Friday, September 09, 2005

Trading Period 2 - More of the Same

It isn't because I have nothing to say that has caused my absence from posting, but rather, I spend a lot of time running my investment business and need to "look" for time to post. Rest assured that I will post my entire story - and some may even find it interesting. In any case, here is the next chapter in a story that will ultimately bring everybody to my current status as director of an investment company. Before I continue, I feel that it is imperative to say that in no way am I trying to solicit interest in my company. It is no acciddent that I have not mentioned my company's name or have given indication of a website. I actually do not need people to invest my company as the returns are sufficient for me to do very well without recruitment. We have approximately 30 investors at the moment recruited through word of mouth only. Another reason why I am not soliciting is a big one - I am prevented from soliciting investors by NASDAQ regulations which state that I can not actively recruit investors as that would mean I am selling securities which is not legal in the context of our business framework. Currently we are a partnership - if and/or when we incorporate, then I will start soliciting! To sum up, as a partnership we can take members through family and word of mouth - but NO advertising.

Back to the story. Our second period of paper trades commenced in late August, 1999 and ended on September, 24th 1999. The following are the results for this trading period:

Average investment per day: $9,629 (this represents $4000 per stock for 24 days @ 3 stocks per day. The reason the number seems odd is because some days only had 1 or 2 candidates)
Total hypothetical profit: $5925.00
Return on initial bank balance of $12,000.00: 49.8%*

Like our first 27 days of trading, this period was fantastic and of course was just additional justification to continue the experiment. Already we were chomping at the bit to dump some real money into this system but we thought that continuing for at least a couple months was prudent, besides, if the system didnt work for the long haul, then it would be considered just another one of those "fluke" systems. Again, one of the first questions people asked was, "Are your returns that high because of market movement"? I looked at the overall market conditions during this period and the following is a summary of both NASDAQ and DOW performance (this can be independently verified on Yahoo):

NASDAQ starting level during this period: 2648 - Ending level: 2740 (3.5% gain)
Dow starting level during this period: 11100 - Ending level: 10280 (7% loss)


As can be seen, neither market was a stellar performer during this period, so market movement was NOT responsible for our outstanding performance. This suggests that for the second month in a row, our hypothesis has held as both markets for these two trading periods are a net negative.


The next chapter in this saga will be "The Culmination of Paper Trades (The Next 4 Months Analysis)"

*This total included commission that would have been charged by Ameritrade ($16.00 total for each in and out - total charges per day were typically $48.00) - Gross return was: 56.8%

Tuesday, June 28, 2005

The End of the First Month

The first 27 days of paper trading was completed in August of 1999. About halfway through this month I started tracking the relative movement of the DOW and NASDAQ and compared it to the daily results of my stock choices. To reiterate, my theory was that investor behavior would be strong enough to be essentially uncorrelated with overall market movement. In other words, even if the markets sold off, the three stock candidates I chose would be unaffected due to predictable investor behavior. The upside to this is that I would not realize tremendous losses just because the overall markets were bad - the downside is that I may not realize accelerated gains just because the market rallies. The point in choosing these candidates was that they were predictable in any given market. After the first month, my theory held firm with one not so surprising adjustment. The performance of my stock candidates was essentially uncorrelated with market movement however, big sell offs did hamper gains. This was fine with me because the candidates were still profitble overall in down markets - I was happy in the knowledge that the market could go to zero and I would have made money overall. This was a tremendous confidence builder for the strategy. There is also some good news - market rallies did in fact accelerate gains. So my first modification to the theory behind investor behavior and the strategy was as follows: Individual investor behavior can indeed be predicted with a fair amount af accuracy, however the performance of the market adjusts the window of profitability. In simpler words, down markets typically temper profit but do not eliminate them, and rallies typically accelerate profit. The end results averages out pretty close to the original hypothesis so there was no reason to change any of the core ideas after month one.

The following are some results for the initial 27 days:

Average investment per day: $10,963.00 (this represents $4000 per stock for 27 days @ 3 stocks per day. The reason the number seems odd is because some days only had 1 or 2 candidates)
Total hypothetical profit: $6438.00
Return on initial bank balance of $12,000.00: 58.7%*

58.7% was a great return for 27 trading days and of course was enough justification for me to paper trade for an additional period of time (next trading period was 24 days). As I stated previously, simple market movement were also investigated for the last 15 trading days of this period. Here are the results of those findings:

NASDAQ starting level during this period: 2840 - Ending level: 2648
Dow starting level during this period: 11200 - Ending level: 11100
Both major indices posted a loss during this 27 day period.


During this period the NASDAQ was down 9 days. On those days, stock candidates produced a gain of: $1850.00
The 6 days that the NASDAQ was positive resulted in a gain of: $1900.00**

The next chapter in this saga will be "Trading Period 2 - More of the Same"

*This total included commission that would have been charged by Ameritrade ($16.00 total for each in and out - total charges per day were typically $48.00) - Gross return was: 68.2%

**The DOW was down on the same days as the NASDAQs and up on the same day so the results would be the same for the DOW performance. One can see that at least for this period of time the correlation between major index performance and stock candidate performance is near zero.

Wednesday, June 22, 2005

Some Concerns

So my first week was good, but was it beginners luck? Could it be that the markets just happened to be good for those first 5 days? Unfortunately, I didnt keep records of the market conditions for about the first two weeks. The main reason I was unconcerned about the overall market performance (NYSE, DOW, NASDAQ, S&P) was that investor behavior for individual stocks chosen was theoretically uncorrelated with the inertia that moves the market indices. I figured that if the indices were globally rallying, my stock candidates would get added incentive to move up. If the indices were flat, my candidates would simple behave in a way congruent with my hypothesis (in other words upward bias). If the indices were in "sell off" mode, it may temper my gains, but in most cases, not be enough to overcome the upward bias. Obviously catastrophic sell offs and certain events (9/11 for example*) would be enough to overcome the positive bias and result in substantial losses. In order to make this as scientific as possible, I decided to record the 3 major market levels on a daily basis and test for correlations between the day's stock performance and market change.

The next chapter to this story will include the first month's results along with a first analysis of market index/stock candidate correlation.

*When markets resumed trading after 9/11 one of the stock candidates opened for a profit - a powerful sign of just how strong individual investment behavior can be

Tuesday, June 21, 2005

The End of Week 1 - Paper Trading Results

The end of week 1 was an eye opening experience. I was very encouraged by the relatively consistant behavior the stocks have shown, and gave me confidence that those variables I thought were important were in fact validated. I fully expected that the variables chosen would result in a bias toward profitable stocks picks, but what I did not expect was that in this first week that bias would have a winning to losing ratio of 11:4 - 73.3% of my stock choices resulted in profits and only 26.7% resulted in loss. These results were obviously very good, but perhaps even better was the fact that they happened the first week of tracking. I would never assume these results would be typical every week (although that would be great) but the fact that they occurred so early gave me a kick start to continue to track. I hate to think what would have happened if the first week was a bust - I may have tried some alteration to the strategy and never realized how good this initial strategy would prove to be. In any case here are some actual numbers for the first week in terms of dollars:

Total stocks bought (21 [3 stocks x 5 days])
Cummulative gain: Approx. $2150.00
Starting balance $12,000.00
One week return: 17.9%

The next time I write, I will discuss the return for the first full month of tracking. I will also discuss some small but important changes made to the strategy. I will also give the details of my first losing day - yes there was a day that I would have lost money, it occurred on my 9th trading day and wouldnt be repeated until the 25th day!

Next Post: Some Concerns and maybe some results

Thursday, April 28, 2005

The First Paper Test

We are still in 1999 when I formed the first formal hypothesis concerning what I believe to be strong psychological influences leading to an investor's decision to buy (and sell for that matter - but that is another story). I wrote my psychological factors down and proceeded to quantify those factors into something I could easily research on the internet. Generally, the hypothesis states that there are stocks that have a "built-in" upward bias. Quantifying this upward bias was not easy, but eventually I settled on a few potential ways to do this. With paper in hand and Yahoo as a stock guide I sat down at the start of July 1999 and at 8:00am CST, I wrote down my first 3 hypothetical stock transactions in excel and thus started 8 months of exhaustive paper trading.

While tracking stocks during the next 8 months, I decided on a hypothetical $4000.00 per stock with a total of 3 stocks per day. I came up with a total of $12,000.00 investment because that is what I had available on my credit card just in case I stumbled on something. Yes, taking money from a credit card is stupid and as I stated in a previous post, I lost a bundle. I am not going to report the actual stocks I purchased as my strategy is now a company secret, but I will give buy prices/ sell prices a dates of executions. My first three paper trades were as follows:

BOT XXXX @ $11.031 and SOLD @ $12.375. A 12% profit approx. $480.00
BOT XXXX @ $37.625 and SOLD @ $38.250. A 2% profit approx. $67.00
BOT XXXX @ $14.250 and SOLD @ $15.000. A 5% profit approx. $210.00

Not bad, I bought $12,000.00 worth of stock and gained about $750.00! Plus I have no overnight exposure - it would all be realized. In any case it was a good start and I was looking forward to day 2. For reference, the NASDAQ was about 2800 at this time and the DOW was at about 11200.

Next Post: The Next Days and the End of Week 1

Wednesday, April 27, 2005

Initial Bad Times

Sometime in 1999 I decided to invest in some stocks. I had no idea how to pick them so I decided to choose a couple of stocks that other people said were "can't misses". I borrowed money from a credit card and opened an E*Trade account with $4000.00 dolars and proceded to buy 20 shares of JDSU at a price of $135.00 per share. I also bought RFMD at a price that escapes me. I decided that a profit was all but given with these 2 stocks and proceeded to work on my other theory while the money rolled in. Fast forward a couple of months and JDSU was at about $100.00 per share and I was out 700 bucks. A while later it was back to my buy point - and this is the part I rememember best. I stopped at my parent's house for a visit and told my dad that JDSU was reporting earnings and we watched it skyrocket right past my buy point into the world of profitability - I was kinda proud. Unfortunately that was my last chance to sell at a profit. Fast forward a year and I sold JDSU for a $2000 loss or what amounts to a 74% loss*. No need to talk about RFMD because the same thing happened there! The lesson I learned was as follows: Never buy a stock for no apparent reason.

I was stunned but was still convinced that a person could predict with accuracy, mass investor behavior if given enough knowledge. I cashed out all my losses and never bought stock again until I started my business. My goal now was to uncover underlying reasons why people buy certain stocks

Next Post: Paper trading with my rudimentary stock strategy

*JDSU is now down 98.5% from my buy point and RFMD is down about 90% - good thing I sold them when I did

Tuesday, April 26, 2005

The Early Years

Back in 1999 I was working on completing my Master's thesis I recruited the help of a research assistant to research literature related to my topic. I apparently interupted him while he was surfing the internet looking up stock quotes on internet related stocks (YHOO, AOL, ASKJ, etc.). I didnt think much of it because I had no interest in the stock market and apparently no work for the RA. Out of curiosity, I asked him if he owned YHOO stock. He said he did and was making a lot of money. This same thing happened over the course of the semester and it seemed everytime I asked him if the stock went up he said yes - of course, in retrospect, every stock was going up relentlessly that year. Over the course of the semester, I started thinking more and more about investing and tested many published theories I heard about(Channeling, Dogs of the DOW, etc.) none of which worked. Yet, this RA guy was making money all the time. I decided to come up with a theory why YHOO and a bunch of other stocks were all going up and up and decided that it actually had little to do with market behavior, company fundamentals and other technical information - but a lot to do with investor behavior. So, my new job from that point was to develop a hypothesis that would allow me to predict individual investor's behavior before it occurred.

Next Post: My first foray into the abyss that was the market sell-off