UQO Working Paper No. 3 — Hedonic housing price models for the US: parametric, quantile, and machine-learning approaches.
1n_trees,r2_geo,r2_eval_within_training_states,r2_random_test225,0.47473288467536123,0.6507742986569474,0.62660341154371350,0.5459552714154834,0.7443231829631414,0.7233258267519742475,0.5498669698126246,0.7721940222677715,0.75329707380786655100,0.5439032982679312,0.7852917594877921,0.76899908683146846150,0.5234242431309662,0.7986188073619156,0.78642629926976837200,0.5108503552703889,0.8051179677128695,0.79492498633519868300,0.487442106425633,0.814065069313564,0.80686241043539949400,0.4653458062209582,0.8195994293676603,0.814356689536608810500,0.436839473621238,0.8236283571397429,0.819147147925790611600,0.43101871774282474,0.8268794034570226,0.82287965016588412700,0.4307539983193228,0.8294270126043002,0.82632687965029413800,0.4284625016184065,0.8313676010395691,0.828922826059054114900,0.42593589924885766,0.8328736764504753,0.8308333173940688151000,0.42501500587502516,0.834330615161805,0.832660931161438716