detect the fraud
3.How to detect the fraud?
There are many ways to detect the fraud .They can simply be divided into two categories
One is quantitative analysis,another is qualitative analysis. In quantitative analysis,the sample is fifty thousand investment company which issued form ADV , form ADV is required to issue to SEC when the investment company's client number more than 15 people or the size of asset under management is larger than $ 25 million , if you did not issue the document or it is not correctly answered or you hide the truth,the SEC will dismissed the document from the investment company which want to register so they can not run their business .In the statistical model, if the investment company in 2001 to 2010 occurs fraud incident ,the dependent variable will equal one.And in a fraud event,the possible property of factor which cause a fraud incident set as dummy independent variables . For example,the document will ask the person in charge whether involved in civil or criminal violation of the law in the past ten years,the dummy variable equal one if the answer is yes otherwise zero; whether the firm’s affiliates has previously been accused of fraud, dummy variable equal one if the answer is yes otherwise zero.Use these more than two thousand samples which occurs fraud incident to do a regression in statistic and do a student-t test. Finally, if these dummy variables are significantly different from zero in the student-t test,we can use these indicators as our guidance,The test results showed that the indicaoors of the company in violation of regulations, civil or criminal violation of the law, the investment company recommend securities in which it has an ownership interest or serves as an underwriters or has any other sales interests, the company employs a registered representatives of a broker-dealer agents , registered investment company act of 1940 were significantly different from zero in a student t-test.So we think these indicator are important
There are many ways to detect the fraud .They can simply be divided into two categories
One is quantitative analysis,another is qualitative analysis. In quantitative analysis,the sample is fifty thousand investment company which issued form ADV , form ADV is required to issue to SEC when the investment company's client number more than 15 people or the size of asset under management is larger than $ 25 million , if you did not issue the document or it is not correctly answered or you hide the truth,the SEC will dismissed the document from the investment company which want to register so they can not run their business .In the statistical model, if the investment company in 2001 to 2010 occurs fraud incident ,the dependent variable will equal one.And in a fraud event,the possible property of factor which cause a fraud incident set as dummy independent variables . For example,the document will ask the person in charge whether involved in civil or criminal violation of the law in the past ten years,the dummy variable equal one if the answer is yes otherwise zero; whether the firm’s affiliates has previously been accused of fraud, dummy variable equal one if the answer is yes otherwise zero.Use these more than two thousand samples which occurs fraud incident to do a regression in statistic and do a student-t test. Finally, if these dummy variables are significantly different from zero in the student-t test,we can use these indicators as our guidance,The test results showed that the indicaoors of the company in violation of regulations, civil or criminal violation of the law, the investment company recommend securities in which it has an ownership interest or serves as an underwriters or has any other sales interests, the company employs a registered representatives of a broker-dealer agents , registered investment company act of 1940 were significantly different from zero in a student t-test.So we think these indicator are important