Abstract: Predicting public company stock upswing and downswing is a very complex and mathematical problem. Accurate prediction of public stock performances can be beneficial to hedge funds, private investors as well as the common public interested in investment in stock.The biotechnology industry has grown rapidly in recent years, doubling in size between 1993 and 1999. 150,800 of the US jobs were generated directly by biotechnology companies, while the remaining 286,600 jobs were generated by companies supplying inputs to the industry, or by companies providing goods and services to biotechnology employees. Biotechnology is a field of science which has enormous potential in growing into a global crisis savior with recent developments ranging from easy surgical techniques and cheap pharmaceuticals to solutions in carcinogenic problems and nanotechnology. Due to the presence of solutions to such immensely concerning humanitarian problems, Biotechnology would soon serve as a backbone to human sustenance and evolution. All this would only mean that the stock prices of such a sector would boom to great heights in the near future. Due to this reason, in depth analysis of stocks of public Biotechnology companies is very important. In this paper, an elegant real-time public stock performance prediction methodology based on an aggregation of historical stock performances and organizational stock properties using a set of top US Biotechnology public companies has been introduced. In this paper, a striking co relation between the stock price and market capitalization swings of the companies with the number of employees has been reported. With these temporal stock swings, repeater operator analysis has been used to establish a statistical veracity of the novel metric of these top biotech companies. This corresponds to a result of profit, 85% of the time. A MATLAB model has also been developed to aid automate the stock performance prediction methodology reported in this. This method lends itself to broaden a large scale statistical validation as well as development of more advanced and complex models for stock performance prediction. For this analysis, exactly 9,234 data points were considered by manual data mining.

Keywords: Earnings per share, P/E Ratio, Shares Outstanding, Market Capitalization, Innovation Potential (IP).