I have the following problem that has to be solved in R ( i need all code).
There is a file with 3 variables a, b, c, and a dependent variable y, all separated by "|" (you can create a file with 100000 rows with randomly generated numbers).
1) Run linear regression producing y' = f(a, b, c).
1a) Save regression formula
2) Replace y with y'
3) Create another file with exponential weighted values of each a, b, c, and y' we will call them aw, bw, cw are weighted using multiple e1 based on previous values, yw is weighted using multiple e2 and is based of forward values of y'.
4) Merge resulting files from steps 2 and 3
5) Run regression, producing field y''. y" = f(a, b, c, aw, bw, cw) y'' should be added as 9th field in file.
5a) Save regression formula
6) Produce statistical analysis of the results
Once you have code for the above,
7) Write steps 1-7 into a formula (this formula will have parameters filename, e1, e2, and will produce r square as the result.
8) Use rmalschains package to produce formula with minimum r square.
9) Produce resulting linear coefficients for a, b, c, aw, bw, cw as well as e1, e2
This should all be done in one routine and I need everything in lightly documented code (ie point out steps 1, 2, etc).
Please let me know if you have any questions.
I work in the financial industry look forward to developing an ongoing relationship with the winning bidder.
Hello, I have done many linear regressions in R. I can also read data from and save data to a file as needed. This sounds like a good opportunity for future work and I am excited to work with you. Robbie