R Script Optimization and enhancement

The objective of the project is to fine tune and enhance an existing R Script that analyses daily sales and returns a simulation of future sales based on historic data. After that, it simulates a loan repayment (cash advance from future sales) based on amortizing a fix proportion of daily sales to the loan. We have an R script that: 1) Loads a file with two variables: "Date" and "Sales". 2) Then, the script generates 2 new variables "Date(1)" and "Sales(1)" for all non-zero Sales values. 3) Next, we make a distribution fit-test for Gamma, Weibull and LogNormal distributions using the "fitdistrplus" package. Results are printed 4) Next, we create 3 simulations (n=1000 trials time=365 days) for "SalesForecast", using the resulting parameters from the Gamma, Weibull and LogNormal distributions obtained from the previous step (simulation values are strictly >= 0). 5) A new variable "Payments" is created based on the resulting "SalesForecast" simulations, which is a fraction of daily "SalesForecast" up to the amount of the debt. Results of simulations are printed. 6) an IRR is calculated using the XIRR() function for each simulation and results are printed OBJECTIVES: a) Modify distribution model to improve simulation prediction (we can show examples of simulation histogram distribution versus historic sample for current model) b) Provided that: - Discount is the percentage discount of the cash advance (i.e. borrower discount future receivables of 10,000 at 20% and gets 8,000 deposited on day =0) - Debt service payment is a fraction of SalesForecast, Then, we would like to plot an (X,Y) chart with a certain level of confidence for an IRR of (z) (e.x. z = 30%) the variables (X=discount, Y=fraction of SalesForecast). If we are satisfied with the results the next step would be to adapt the script to do the same for a large number of potential borrowers (one alternative is to load 1 file with 3 variables, "Borrower-id", "Date" and "Sales") and loop the process for each unique "Borrower-id".