Our models were very accurate in predicting the performance of small business entrepreneurs

On Monday, after about an hour of attorney type of hassling, we agreed on terms (which I’d failed to specify in my vision and weren’t as good as I would have liked) and my attorney drafted a 1.5 page letter agreement that we both signed. My attorney then pulled an unused corporation out of his files, we decided on a name, he contacted the Attorney General of Maryland to change the corporation’s name to our new one, we set up a bank account and two days later we had the $2.5 million in our account via wire transfer.

That was my first big success using what evolved into the Cybernetic Transposition techniques.

In any case, I decided I wanted to do my dissertation on the unconscious decision making of venture capitalists. To do so, I had to develop a system that would model that decision making. Thus was born my Arintel system. It used artificial intelligence pattern recognition techniques to create an imaginary 15 dimension “hyperspace” in which people with similar unconscious personalities would cluster together.

Don’t worry if that doesn’t make any sense. Arintel isn’t what you will learn in this book. You’re going to learn something even better.

Arintel worked fabulously even though the data collection was a bear and the program would only run on the most powerful supercomputer of the time, a Cray I, hogging expensive computer time like a demon. In 1973 dollars, it cost $25,000 to make one run.

I then applied Arintel to sophisticated market and economic research on behalf of the U.S. government and Fortune 500 companies. (They were the only ones willing to spend $500,000 to $1 million per project.)

In doing research in many parts of the world, I found some very interesting things. The most interesting was that we always found the same 12 clusters of unconscious personality types. That was so strange that I brought in experts on statistical techniques to see if that was a result of bias in Artintel. They concluded it wasn’t.

In one project, we brought together people who were most typical of the most interesting clusters (stereotypes) and found that behaviorally and in terms of mannerisms, these people from all over the world were almost clones of each other.

The models Artintel developed were so powerful that, in doing a 7-year study of the operations of the U.S. Small Business Administration, our seven variable models were 90 to 99 percent accurate in predicting the performance of small business entrepreneurs.

to be continued…

Tags: , , , , , , , , , , ,

Similar articles

Comments are closed

Additional Articles From "Entrepreneurship"