7,649
edits
Changes
Jump to navigation
Jump to search
Urban Start-up Agglomeration and Venture Capital Investment (view source)
Revision as of 11:36, 10 September 2019
, 11:36, 10 September 2019→Maximum R-squared
So, aside from the Palo Alto result, this is clearly a greatly preferred spec. I think N>=12 (just over half of the 23 years) is fair as a cut off. Also Portland, OR, maximizes at 4 hulls on both measures with N=15...
The only thing that we should change is the R2 estimation regression. Up until this point, we've been using:
pca nosinglemulti nopair nohull totsinglemulticount totpaircount tothullcount totpairlength tothullarea
predict pc1 pc2 pc3, score
quietly capture reg growthinv17lf pc1 pc2 pc3 if placeid==`placeid' & numclusters==`clusters' & lowesthighestflag==1 & year>=1995 & year <.
There are two issues. Why are we using a pca? Just to get the number of regressors down? And more importantly, one or more PCA components may be picking up a scale effect. We don't want to use the scale regressors in R2 estimation, because they might drive the R2. '''So we should, for each regression or beforehand, create residuals from estimating the scale effect and then use then as our dependent variable.'''
=====Revisiting Portland=====
The resulting map has much more adjacency than overlap. Measuring the nearest hull edge and center distance for each hull in a year to each hull in the next year and averaging would compute two measures of hull persistence. The overlap area from year to year, either in total or as a fraction of the second year's (or smaller years) total area, would provide another measure of persistence.
=====What do we want to know?=====
So now we have 200 (ish) cities with their optimally selected hulls (we chose the best hull count that is constant from 1995 to 2017 using the lowest-highest occurrence of that count). And now we'd like to know:
*Whether having fewer hulls is associated with growth, controlling for size -- it is: nohulll -.168335***
*Whether having a greater hull density is associated with growth, controlling for size -- it is: tothulldensityl .0730263***
*Whether having a higher fraction of locations inside hulls is associated with growth, controlling for size: -- it isn't: frachull -.1345406*
*Whether having hulls closer together is associated with growth, controlling for size. '''We should put these layer in the list to build avghulldisthm and avgdisthm (see line 1335).'''
=====Houston, TX=====
We also want to know about Houston, TX.
tab place placeid if placeid >50 & placeid <100
//Houston is 83
tab chosenhullspcar2inc if placeid==83
//10
SELECT year, layer FROM MasterLayers
WHERE place='Houston' AND statecode='TX' AND layer=lowesthighestlayer AND numclusters=10
ORDER BY year, layer;
1990 20
2000 45
2005 50
2010 30
print myDict["Houston_TX"];
[-95.836717, -95.014608, 29.515925, 30.155908]
===Image Analysis===
