Predictor sets
Predictors sets are sets of predictors (independent variables) seen by models.
Changing predictor sets across models allows us to study how predictor selection affects model performance and biases.
Predictor set specifications
The below table contains the full list of predictor sets used in model specifications.
For an interpretation of the meaning of each variable, see the full list of predictors.
Predictors in the column continuous are used as is. Predictors in dummied are first transformed into categorical dummies (e.g., for year-quarter or regional fixed effects). Stacked (second-level) models use the predictions of first-level models as predictors: lnusd-ha_<model identifier>.
label |
continuous |
dummied |
|
X |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-soilif |
Main + soil “if” |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_primeif, f_soil_statewide, f_soil_statewideif, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-soildetail |
Main + all soils |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_primedrain, f_soil_primeirrigate, f_soil_primeprotect, f_soil_statewide, f_soil_statewidedrain, f_soil_statewideirrigate, f_soil_statewideprotect, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-nosoilclim |
Main - climate/soil |
cst_2500, cst_50, elev, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-pnas |
Nolte 2020 PNAS |
cst_2500, cst_50, elev, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lat_id, lat_id_rot45, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, slope, travel, water_exposure, year_cont |
|
X-irr |
Main + irrigation |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, irr_2000_2020, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-bare |
Nothing |
lat_id, lat_id_rot45, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, year_cont |
|
X-people |
People |
lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, year_cont |
|
X-buildings |
Buildings |
lat_id, lat_id_rot45, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_bld_fp_500, p_bld_fp_5000, year_cont |
|
X-wealth |
Wealth |
hh_inc_med_bg_2012-2016, lat_id, lat_id_rot45, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, year_cont |
|
X-peoplewealth |
People & wealth |
hh_inc_med_bg_2012-2016, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, year_cont |
|
X-peoplebuildings |
People & buildings |
lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_bld_fp_500, p_bld_fp_5000, year_cont |
|
X-popbldwealth |
People, buildings, wealth |
hh_inc_med_bg_2012-2016, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_bld_fp_500, p_bld_fp_5000, year_cont |
|
X-fp |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, n_bld_fp_per_ha, p_barren, p_bld_fp, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-latlong |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, ln_bld_pop_exp_c4, ln_ha, long_id, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-nonspatial |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, ln_bld_pop_exp_c4, ln_ha, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-xy45 |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, ln_bld_pop_exp_c4, ln_ha, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, x, x45, y, y45, year_cont |
|
X-blackhispanic |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_black_bg_2012-2016, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_hispanic_bg_2012-2016, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-diverse |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_asian_bg_2012-2016, p_barren, p_black_bg_2012-2016, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_hispanic_bg_2012-2016, p_mixed_bg_2012-2016, p_native_bg_2012-2016, p_pacific_bg_2012-2016, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-yha |
Main predictor set, yha |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-yhac |
Main predictor set, yhac |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel, year_cont |
|
X-yqc |
Main predictor set |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, hh_inc_med_bg_2012-2016, lake_dist_asinh, lake_exposure, lat_id, lat_id_rot45, ln_bld_pop_exp_c4, ln_ha, long_id, long_id_rot45, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel |
|
X-main-lm_latlong_Xha^2_Xyqd |
Main, curvature, dummies, Xha^2 |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, lake_dist_asinh, lake_exposure, lat_id, ln_bld_pop_exp_c4, ln_bld_pop_exp_c4^2, ln_ha, ln_ha^2, ln_hh_inc_med_bg_2012-2016, ln_hh_inc_med_bg_2012-2016^2, long_id, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel |
year_quarter |
X-main-lm_latlong_Xha^2_Xyrd |
Main, curvature, dummies, Xha^2 |
clim_ppt_summer, clim_ppt_winter, clim_tmean_summer, clim_tmean_winter, cst_2500, cst_50, elev, f_soil_farmlandofuniqueimportance, f_soil_local, f_soil_prime, f_soil_statewide, fld_fr_fath_f100, fld_fr_fath_p100, lake_dist_asinh, lake_exposure, lat_id, ln_bld_pop_exp_c4, ln_bld_pop_exp_c4^2, ln_ha, ln_ha^2, ln_hh_inc_med_bg_2012-2016, ln_hh_inc_med_bg_2012-2016^2, long_id, m2_bld_fp_ihs, n_bld_fp_ihs, p_barren, p_bld_fp_500, p_bld_fp_5000, p_crops, p_forest, p_grassland, p_pasture, p_prot_2010_5000, p_shrub, p_wet, rd_dist_pvd+, river_exposure, slope, travel |
year |
stack1 |
Stack1 |
lnusd-ha_region-nb, lnusd-ha_region-nb_lm, lnusd-ha_region-nb_hgb, lnusd-ha_region-nb_mv, lnusd-ha_region, lnusd-ha_county_lm, ln_ha, lat_id, long_id, lat_id_rot45, long_id_rot45, year_cont |
|
stack2 |
Stack2 |
lnusd-ha_region-nb, lnusd-ha_region-nb_lm, lnusd-ha_region-nb_hgb, lnusd-ha_region-nb_bare, lnusd-ha_region-nb_mv, lnusd-ha_region-nb_mv_lm, lnusd-ha_conus_lm_Xyqd_high, ln_ha, lat_id, long_id, lat_id_rot45, long_id_rot45, year_cont |