Forum Discussion
column data linear fit the linear formula Vs = A +B*Vp -C*NPHI - D*RHOB
A little Help please,
I have column data of the inputs Vp, NPHI and RHOB vs depth and wish to find best fit to Vs also on same depth index? The starting point of the formula would be Vs = 1.234 + 0.508*Vp - 0.235*NPHI - 0.074*RHOB and would like to vary all variables A,B,C,D +/- some value like 2.0 hoping that the equation is a good start for linear fit. I don't expect someone to do all my work just get me started somewhere until I run into problems. I have added solver and crystal ball to my Excel thinking a Monte Carlo analysis would be cool or if the solver alone would be able to find best fit for the variables?? I do want to try and stay with the format of original equation. I have maybe 7000 feet of Vp,NPHI,RHOB, and Vs data to run through the curve fitting to test each model. If I find a good fit on this set of data then I will try and use it as a predictor of Vs based on Vp for similar cases near this location.
Thanks,
Tom_SIII
1 Reply
Tom_SIIImaybe I'm too simplistic here but I assume you have 3 columns of data that correspond to Vp, NPHI, and RHOB, so I would add 4 rows to the top of the data for parameters A,B,C,D and then make the formula in column 4 (starting in row 5) =d$1+$a5*d$2-$b5*d$3-$c5*d$4
the strategic use of $ to fix either the row or column makes so you can fill down to make the calculation for every row. Then you can go to the next column and adjust A,B,C,D (manually or using a formula) and fill right to calculate the values with the new A,B,C,D.
hope that helps
