{"id":"W3047241722","doi":"10.1162/rest_a_01301","title":"Improving Estimates of Transitions from Satellite Data: A Hidden Markov Model Approach","year":2023,"lang":"en","type":"article","venue":"The Review of Economics and Statistics","topic":"Atmospheric and Environmental Gas Dynamics","field":"Environmental Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Ground truth; Satellite; Deforestation (computer science); Econometrics; Computer science; Markov chain; Markov model; Observational error; Satellite imagery; Transition (genetics); Hidden Markov model; Remote sensing; Algorithm; Artificial intelligence; Mathematics; Geography; Machine learning","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006895536,0.0006898225,0.0009786973,0.002410094,0.0004783941,0.001418047,0.001588592,0.0009134201,0.001341018],"category_scores_gemma":[0.0270074,0.0007555593,0.001103497,0.001807963,0.0008046926,0.00178321,0.001352634,0.002278447,0.0003239407],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001368353,"about_ca_system_score_gemma":0.001212583,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01984434,"about_ca_topic_score_gemma":0.01713126,"domain_scores_codex":[0.9981039,0.001070941,0.0001105913,0.000413276,0.0001982786,0.0001031599],"domain_scores_gemma":[0.9782791,0.01796625,0.001529286,0.001068838,0.001005319,0.0001513564],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001461351,0.00009351337,0.04716472,0.0002414633,0.0005416422,0.0001571819,0.0002663755,0.7969028,0.0007913163,0.04869812,0.002868406,0.1021284],"study_design_scores_gemma":[0.000008256926,0.00001312791,0.003998291,0.00004982979,0.00004152301,0.00001735969,0.00002426309,0.9586214,0.0003863398,0.03612046,0.0006982156,0.00002096837],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05327756,0.0008521218,0.9431109,0.0007951281,0.00006849854,0.0000368559,0.0006578816,0.0003655464,0.0008353598],"genre_scores_gemma":[0.844178,0.001044368,0.1507335,0.0001937596,0.0001392227,0.0001028637,0.002011762,0.0001223388,0.001474286],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01984434,"threshold_uncertainty_score":0.03945768,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02302391707634279,"score_gpt":0.2331119199307656,"score_spread":0.2100880028544228,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}