Comment on Why 30.1% of numbers start with 1Comments−jacquesm16yBenfords law is instrumental in fraud detection, the Enron guys were wise to it, but Bernie Madoff was not:http://falkenblog.blogspot.com/2008/12/benfords-law-catches-...−wooby16yAs a friend of mine pointed out, how is this possible? Shouldn't fraudulent numbers also follow Benford's law?−count16yIf you are aware of it, you make sure your fake data follows Benford's law. Otherwise, your fake data will be 'random', because people assume that is how it is supposed to look.−sown16yhttp://news.ycombinator.com/item?id=1076534−schammy16yRandom data doesn't follow the law. Fraudulent tax returns are essentially random numbers, hence, detectable.−shrughes16yUniformly distributed random data doesn't follow the law. Data whose logarithm is uniformly distributed does.
Comments
Benfords law is instrumental in fraud detection, the Enron guys were wise to it, but Bernie Madoff was not:
http://falkenblog.blogspot.com/2008/12/benfords-law-catches-...
As a friend of mine pointed out, how is this possible? Shouldn't fraudulent numbers also follow Benford's law?
If you are aware of it, you make sure your fake data follows Benford's law. Otherwise, your fake data will be 'random', because people assume that is how it is supposed to look.
http://news.ycombinator.com/item?id=1076534
Random data doesn't follow the law. Fraudulent tax returns are essentially random numbers, hence, detectable.
Uniformly distributed random data doesn't follow the law. Data whose logarithm is uniformly distributed does.