Build a PC app to generate multivariate normal variables from uncorrelated random number input.

I would like a PC based app to allow me to input a file with uncorrelated random numbers for several variables and input a file with a set of correlations, and have the app generate a multivariate normal distribution such that the several variables have the intercorrelations that I specify. In random number inputs can be one of two types: normally distributed or uniformly distributed. Attached are two small files that contain samples of the random number input to the app and a third file with an example of the intercorrelations required: 1. The file [login to view URL] contains 2,000 lines of gaussian random numbers for 5 variables 2. The file [login to view URL] contains 2,000 lines of uniformly distributed random numbers for 5 variables 3. The file [login to view URL] contains a sample specification for the intercorrelations I want in the multivariate output file. This xls file also specifies the nature of the input (uniform or normal) and the number of variables and input data lines. The number of variables might be as great as 10. The number of data lines may be 20 million. As part of the deliverables, I would like a description of the mathematics that the program uses. I would like the app to output the multivariate normal variables in an CSV file similar to the random number input file and with the same file name but with the extension OUT rather than TXT. I would like to specify the names of the two input files: one with random numbers and one with specified intercorrlations. These web pages seem to address some of the statistical approaches to intercorrelation aspect of this project: [login to view URL] [login to view URL] [login to view URL] This is not a big project. There is no fixed due date, but it should not take long. A week should be long enough.