7,612
edits
Changes
Jump to navigation
Jump to search
We need to fix it in I fixed the following, and rerun reran the analysisusing fracunclustered = (layer-1)/(finallayer-1):*elbowcalc->heurcalc*elbowdata->heurdata*Elbowflh-heurflh Plotting the data from heurdata.txt in excel gives: 2.5944 x^3 - 2.9174 x^2 - 0.6407 x + 1.0304Wolfram alpha[https://www.wolframalpha.com/input/?i=inflection+points+2.5944x3+-+2.9174x2++-+0.6407x+%2B+1.0304] says the inflection point is now: x≈0.374833, y≈0.516982 Note that this is now the fraction complete is relative to finallayer-1. For reference, the old value was 0.429388 relative to finallayer.
Urban Start-up Agglomeration and Venture Capital Investment (view source)
Revision as of 19:08, 28 October 2020
, 19:08, 28 October 2020→New implementations
#The maximum R2 method (now just 8 metadata vars)
#%Complete - now i-1/I-1
=====Max R2=====
I ran the code and pushed the results through the database. The newly selected layers are generally the same or one lower that the previously selected layers. I expect that this will be just fine.
=====Fraction Complete=====
gen fracunclustered=layer/finallayer
====Implementing The Elbow Method====
