@RobertEG wrote:
The ability of Fair Isaac to keep the workings of its scoring algorithms as a proprietary trade secret is at the heart of U.S. intellectual property law, and wont change.
When someone invents a product, process, or method of doing business, they have two basic options to protect it in the marketplace. They can get a patent, or they can choose to protect it as a trade secret.
A patent has advantages and disadvantages. To obtain a patent, the inventor must provide a full written dsclosure of how the invntion is both made and used in sufficient detail for anyone to make and use it themselves. When the application is filed, this disclosure is made public 18-months after filing. In exchange for this public disclosure, the patentee is giving exclusive rights to exclude all others from making, using, or selling the patented invention for a non-renewable period of 20-years from date of filing of the application for patent. After that, all exclusive rights are gone, and the invention is in the public domain for anyone to make, use, or sell.
A trade secret has separate advantages and disadvantages. Trade secrets convey no exclusive rights, so are only beneficial if the invention can be kept secret, and also is of a nature that does not permit it to readily be reverse-engineered. If someone figures it out on their own, they are free to make, use and sell it.
Trade secrets have no date of expiration, so can far outlive a patent grant.
Fair Isaac has chosen the trade-secret route. They are banking on being able to keep their algorithms secret and that they wont be reverse-engineered.
It is not surprising that, in order to protect their business product from copying, they are very tight lipped on the precise workings of its algorithms. To expect Congress to require any detailed disclosure of its workings would be totally contrary to 250-years of U.S. intellectual property law.
Exactly so. What many people don't realize about patents is they are fundamentally rewards for disclosing inventions; they exist mainly because many inventions from Ancient and Early Modern times were forgotten because the inventor kept them secret. As a pharmaceutical researcher, I work in a world where patents are of enormous importance: it is nearly impossible to keep the chemical composition of a drug secret, especially since regulators won't approve any drug for sale without a full details about what it contains, so patents are the only way to protect new drug discoveries, and without some kind of intellectual property protection companies like mine couldn't spend the billions of dollars we spend on clinical trials.
As for what BeamMeUp said about comparing FICO and Vantage scores, it is indeed true that because they are different scoring algorithms neither can be calculated very accurately from the other. However, since both are intended to predict the same real-world event (a person having trouble paying bills) among the same population (US consumers), there must be a strong positive correlation between them if both are valid predictors. A person who is at, say, the 75th percential of the population with one scoring model might not be exactly at the 75th percentile in the other model, maybe they are at the 70th percential or the 80th percentile with the other model, but unless that person happens to be in some manner highly unusual statistically speaking it is not likely somebody who is at the 75th percentile in one model would be at the 50th percentile in the other model. It is rather like different college instructors assigning grades in a course with multiple sections (once upon a time I was one of those instructors, so I know how hard it is for a group of instructors to make the grading as fair as possible, and I took part in many discussions of grading standards among the instructors because we truly wanted to be as fair as we could): a student who gets a B from one instructor might have been given a B-plus or a B-minus by another instructor, but it is highly unlikely that student would have been given an A-plus or a C-minus by any of the instructors.
TU 791 02/11/2013, EQ 800 1/29/2011 , EX Plus FAKO 812, EX Vantage Score 955 3/19/2010 wife's EQ 9/23/2009 803
EX always was my highest when we could pull all three
Always remember: big print giveth, small print taketh away
If you dunno what tanstaafl means you must Google it
