Threshold detection and analysis: MARS methodology in Macroeconomics and Public Health

Estás aquí:
  1. Home
  2. Threshold detection and analysis: MARS methodology in Macroeconomics and Public Health
  • Autor: César Nebot Monferrer
  • Director/es: José García Solanes, Arielle Beyaert Stevens
  • Defensa: 30/11/2018 - Universidad de Murcia
  • Tribunal: Gabriel Pérez Quirós, Antonio Montañés Bernal, Vicente Orts Ríos
  • Calificación: Sobresaliente cum laude
  • Ver publicaciones relacionadas

RESUMEN

En esta tesis empleamos y desarrollamos una metodología altamente flexible para detectar umbrales en relaciones macroeconómicas como la Regla de Taylor y la Ley de Okun. Además la aplicamos de forma interdisciplinar en una de las preocupaciones actuales más serias en Salud Pública: la resistencia bacteriana a los antibióticos. Uno de los objetivos es aportar estimaciones que alienten el debate para orientar el diseño de políticas públicas, macroeconómicas como sanitarias.
Tanto estas relaciones macroeconómicas como la resistencia bacteriana a los antibióticos muestran evidencia de no linealidad y posible existencia de umbrales. La metodología Multiple Adaptive Regression Splines (MARS) de Friedman permite detectar estos comportamientos dejando que los datos seleccionen el modelo óptimo y lo estime. La flexibilidad y la capacidad de anidación de MARS no sólo le permiten estimar múltiples umbrales sin imponer previamente qué variables los contienen y cuántos umbrales presenta, sino que además permite estimarlos en efectos de interacción. Describimos en profundidad los algoritmos iterativos que componen MARS para realizar las estimaciones evitando problemas de concurvidad y complejidad innecesaria del modelo y desarrollamos un protocolo de implementación que se adecúa a la naturaleza de los datos y de las relaciones estudiadas. A su vez aportamos una importante innovación metodológica: un procedimiento de estimación de los intervalos de confianza de los umbrales detectados con MARS.
Para la regla de Taylor, desarrollamos un modelo teórico que pone de manifiesto la posible existencia de umbrales en la regla de Taylor. Estimamos la regla de Taylor para la Reserva Federal de Estados Unidos y detectamos umbrales tanto en la inflación como en las desviaciones del output y en la interacción entre ambos, además de un umbral en el efecto de la presidencia de Volcker. Esta caracterización más completa de la Regla de Taylor en sí constituye otra de las innovaciones de esta tesis. Además, estimamos la regla de Taylor para el BCE en la que detectamos umbrales tanto en las desviaciones del producto como en la tasa de inflación y aportamos sus respectivos intervalos de confianza. Nuestro estudio permite poner de manifiesto que el BCE tiene en cuenta la actividad económica más de lo que oficialmente declara y reacciona de forma más intensa a las desviaciones de la inflación cuando exceden con suficiente margen el objetivo oficial del 2%.
Para la Ley de Okun, proponemos un modelo teórico sencillo que permite dar fundamento a la no linealidad observada con umbrales. Además, abre el debate a considerar que su forma se debe a la combinación de la hipótesis de aversión al riesgo de las empresas, la de rigidez institucional y la del acaparamiento de trabajo, en lugar de limitarse a que una se imponga excluyendo a las demás. Estimamos la Ley de Okun para cuatro países de Europa: España, Alemania, los Países Bajos y Francia, controlando además el posible efecto de la crisis en la Eurozona. A pesar de no imponer un número preestablecido de umbrales, nuestros resultados avalan la existencia de un único umbral para cada país, pero en posiciones significativamente distintas entre sí. Por un lado, la Ley de Okun en España muestra pendientes mucho más acusadas que las de los países centrales. Las diferencias entre los coeficientes por debajo y por encima del umbral son consistentes con la hipótesis de aversión al riesgo de las empresas. Asimismo, las diferencias significativas entre las posiciones de los umbrales contribuyen al debate de que la combinación de la aversión al riesgo con la rigidez institucional tiene más fuerza en España que en el resto de países analizados, donde el acaparamiento laboral parece más importante. A la luz de todos estos resultados, nuestro análisis aporta innovaciones tanto teóricas como empíricas en el debate de la Ley de Okun.
Finalmente, ante la preocupante amenaza para la Salud Pública que entraña la resistencia bacteriana a los antibióticos, el interés actual, su no linealidad y posibles umbrales, estimamos estas relaciones en cinco centros hospitalarios europeos en un estudio multidisciplinar de investigadores internacionales. Adaptamos el protocolo de estimación para series temporales diferentes a las macroeconómicas y contribuimos a que la detección de umbrales en la intensidad de uso de antibióticos ayude a orientar el diseño de políticas de Salud Pública. Partiendo de la hipótesis de Levy sobre que la resistencia se asocia a que la intensidad de uso de antibióticos supere ciertos umbrales, nuestras estimaciones detectan estos umbrales en los centros hospitalarios y de atención primaria objeto de estudio. Además aportamos intervalos de confianza que facilitan soporte significativo a las recomendaciones de política sanitaria sobre el uso restringido de antibióticos. Esto representa en si una innovación metodológica fundamental.
Deseamos que estos resultados puedan contribuir humildemente a arrojar algo de luz en el diseño de políticas públicas para afrontar los desafíos actuales que se nos presentan en Macroeconomía y Salud Pública.

Publicado el
WP_Query Object ( [query] => Array ( [post_type] => publicaciones [meta_key] => _wpcf_belongs_tesis_id [meta_value] => 2441 ) [query_vars] => Array ( [post_type] => publicaciones [meta_key] => _wpcf_belongs_tesis_id [meta_value] => 2441 [error] => [m] => [p] => 0 [post_parent] => [subpost] => [subpost_id] => [attachment] => [attachment_id] => 0 [name] => [pagename] => [page_id] => 0 [second] => [minute] => [hour] => [day] => 0 [monthnum] => 0 [year] => 0 [w] => 0 [category_name] => [tag] => [cat] => [tag_id] => [author] => [author_name] => [feed] => [tb] => [paged] => 0 [preview] => [s] => [sentence] => [title] => [fields] => [menu_order] => [embed] => [category__in] => Array ( ) [category__not_in] => Array ( ) [category__and] => Array ( ) [post__in] => Array ( ) [post__not_in] => Array ( ) [post_name__in] => Array ( ) [tag__in] => Array ( ) [tag__not_in] => Array ( ) [tag__and] => Array ( ) [tag_slug__in] => Array ( ) [tag_slug__and] => Array ( ) [post_parent__in] => Array ( ) [post_parent__not_in] => Array ( ) [author__in] => Array ( ) [author__not_in] => Array ( ) [ignore_sticky_posts] => [suppress_filters] => [cache_results] => 1 [update_post_term_cache] => 1 [lazy_load_term_meta] => 1 [update_post_meta_cache] => 1 [posts_per_page] => 10 [nopaging] => [comments_per_page] => 50 [no_found_rows] => [order] => DESC ) [tax_query] => WP_Tax_Query Object ( [queries] => Array ( ) [relation] => AND [table_aliases:protected] => Array ( ) [queried_terms] => Array ( ) [primary_table] => wp_posts [primary_id_column] => ID ) [meta_query] => WP_Meta_Query Object ( [queries] => Array ( [0] => Array ( [key] => _wpcf_belongs_tesis_id [value] => 2441 ) [relation] => OR ) [relation] => AND [meta_table] => wp_postmeta [meta_id_column] => post_id [primary_table] => wp_posts [primary_id_column] => ID [table_aliases:protected] => Array ( [0] => wp_postmeta ) [clauses:protected] => Array ( [wp_postmeta] => Array ( [key] => _wpcf_belongs_tesis_id [value] => 2441 [compare] => = [compare_key] => = [alias] => wp_postmeta [cast] => CHAR ) ) [has_or_relation:protected] => ) [date_query] => [request] => SELECT SQL_CALC_FOUND_ROWS wp_posts.ID FROM wp_posts INNER JOIN wp_postmeta ON ( wp_posts.ID = wp_postmeta.post_id ) WHERE 1=1 AND ( ( wp_postmeta.meta_key = '_wpcf_belongs_tesis_id' AND wp_postmeta.meta_value = '2441' ) ) AND ((wp_posts.post_type = 'publicaciones' AND (wp_posts.post_status = 'publish'))) GROUP BY wp_posts.ID ORDER BY wp_posts.post_date DESC LIMIT 0, 10 [posts] => Array ( [0] => WP_Post Object ( [ID] => 2654 [post_author] => 19 [post_date] => 2019-06-03 12:14:08 [post_date_gmt] => 2019-06-03 10:14:08 [post_content] => Balancing access to antibiotics with the control of antibiotic resistance is a global public health priority. At present, antibiotic stewardship is informed by a ‘use it and lose it’ principle, in which antibiotic use by the population is linearly related to resistance rates. However, theoretical and mathematical models suggest that use–resistance relationships are nonlinear. One explanation for this is that resistance genes are commonly associated with ‘fitness costs’ that impair the replication or transmissibility of the pathogen. Therefore, resistant genes and pathogens may only gain a survival advantage where antibiotic selection pressures exceed critical thresholds. These thresholds may provide quantitative targets for stewardship—optimizing the control of resistance while avoiding over-restriction of antibiotics. Here, we evaluated the generalizability of a nonlinear time-series analysis approach for identifying thresholds using historical prescribing and microbiological data from five populations in Europe. We identified minimum thresholds in temporal relationships between the use of selected antibiotics and incidence rates of carbapenem-resistant Acinetobacter baumannii (Hungary), extended-spectrum β-lactamase-producing Escherichia coli (Spain), cefepime-resistant E. coli (Spain), gentamicin-resistant Pseudomonas aeruginosa (France) and methicillin-resistant Staphylococcus aureus (Northern Ireland) in different epidemiological phases. Using routinely generated data, our approach can identify context-specific quantitative targets for rationalizing population antibiotic use and controlling resistance. Prospective intervention studies that restrict antibiotic consumption are needed to validate these thresholds. [post_title] => A nonlinear time-series analysis approach to identify thresholds in associations between population antibiotic use and rates of resistance [post_excerpt] => [post_status] => publish [comment_status] => open [ping_status] => open [post_password] => [post_name] => a-nonlinear-time-series-analysis-approach-to-identify-thresholds-in-associations-between-population-antibiotic-use-and-rates-of-resistance [to_ping] => [pinged] => [post_modified] => 2021-07-28 10:53:18 [post_modified_gmt] => 2021-07-28 08:53:18 [post_content_filtered] => [post_parent] => 0 [guid] => http://doctoradodecide.com/?post_type=publicaciones&p=2654 [menu_order] => 0 [post_type] => publicaciones [post_mime_type] => [comment_count] => 0 [filter] => raw ) [1] => WP_Post Object ( [ID] => 2584 [post_author] => 19 [post_date] => 2019-05-03 12:11:34 [post_date_gmt] => 2019-05-03 10:11:34 [post_content] => We apply a non-parametric procedure, Multiple Adaptive Regression Splines (MARS), to endogenously select the best multiple threshold model for the European Central Bank (ECB) Taylor Rule (TR). MARS offers the advantage of not excluding simpler models if they better fit the data. On monthly data from 1/2000 to 9/2016, the TR exhibits thresholds for both the output gap and inflation and uncover interesting information about the different sensitivities of the ECB to each variable. We conclude that the ECB cares more about the economic activity than officially declared and reacts to inflation deviations only if they sufficiently exceed the official 2% target. We also detect an overall shift in the monetary policy during the 2008 financial crisis that pushed the system towards the Zero Lower Bound (ZLB). We reach a more complete description of the ECB TR than published so far. [post_title] => New insights into the nonlinearity of the ECB Taylor Rule [post_excerpt] => [post_status] => publish [comment_status] => open [ping_status] => open [post_password] => [post_name] => new-insights-into-the-nonlinearity-of-the-ecb-taylor-rule [to_ping] => [pinged] => [post_modified] => 2021-07-28 10:53:18 [post_modified_gmt] => 2021-07-28 08:53:18 [post_content_filtered] => [post_parent] => 0 [guid] => http://doctoradodecide.com/?post_type=publicaciones&p=2584 [menu_order] => 0 [post_type] => publicaciones [post_mime_type] => [comment_count] => 0 [filter] => raw ) ) [post_count] => 2 [current_post] => -1 [in_the_loop] => [post] => WP_Post Object ( [ID] => 2654 [post_author] => 19 [post_date] => 2019-06-03 12:14:08 [post_date_gmt] => 2019-06-03 10:14:08 [post_content] => Balancing access to antibiotics with the control of antibiotic resistance is a global public health priority. At present, antibiotic stewardship is informed by a ‘use it and lose it’ principle, in which antibiotic use by the population is linearly related to resistance rates. However, theoretical and mathematical models suggest that use–resistance relationships are nonlinear. One explanation for this is that resistance genes are commonly associated with ‘fitness costs’ that impair the replication or transmissibility of the pathogen. Therefore, resistant genes and pathogens may only gain a survival advantage where antibiotic selection pressures exceed critical thresholds. These thresholds may provide quantitative targets for stewardship—optimizing the control of resistance while avoiding over-restriction of antibiotics. Here, we evaluated the generalizability of a nonlinear time-series analysis approach for identifying thresholds using historical prescribing and microbiological data from five populations in Europe. We identified minimum thresholds in temporal relationships between the use of selected antibiotics and incidence rates of carbapenem-resistant Acinetobacter baumannii (Hungary), extended-spectrum β-lactamase-producing Escherichia coli (Spain), cefepime-resistant E. coli (Spain), gentamicin-resistant Pseudomonas aeruginosa (France) and methicillin-resistant Staphylococcus aureus (Northern Ireland) in different epidemiological phases. Using routinely generated data, our approach can identify context-specific quantitative targets for rationalizing population antibiotic use and controlling resistance. Prospective intervention studies that restrict antibiotic consumption are needed to validate these thresholds. [post_title] => A nonlinear time-series analysis approach to identify thresholds in associations between population antibiotic use and rates of resistance [post_excerpt] => [post_status] => publish [comment_status] => open [ping_status] => open [post_password] => [post_name] => a-nonlinear-time-series-analysis-approach-to-identify-thresholds-in-associations-between-population-antibiotic-use-and-rates-of-resistance [to_ping] => [pinged] => [post_modified] => 2021-07-28 10:53:18 [post_modified_gmt] => 2021-07-28 08:53:18 [post_content_filtered] => [post_parent] => 0 [guid] => http://doctoradodecide.com/?post_type=publicaciones&p=2654 [menu_order] => 0 [post_type] => publicaciones [post_mime_type] => [comment_count] => 0 [filter] => raw ) [comment_count] => 0 [current_comment] => -1 [found_posts] => 2 [max_num_pages] => 1 [max_num_comment_pages] => 0 [is_single] => [is_preview] => [is_page] => [is_archive] => 1 [is_date] => [is_year] => [is_month] => [is_day] => [is_time] => [is_author] => [is_category] => [is_tag] => [is_tax] => [is_search] => [is_feed] => [is_comment_feed] => [is_trackback] => [is_home] => [is_privacy_policy] => [is_404] => [is_embed] => [is_paged] => [is_admin] => [is_attachment] => [is_singular] => [is_robots] => [is_favicon] => [is_posts_page] => [is_post_type_archive] => 1 [query_vars_hash:WP_Query:private] => 3d3c4518dfe21505e4dab58cd5680551 [query_vars_changed:WP_Query:private] => [thumbnails_cached] => [allow_query_attachment_by_filename:protected] => [stopwords:WP_Query:private] => [compat_fields:WP_Query:private] => Array ( [0] => query_vars_hash [1] => query_vars_changed ) [compat_methods:WP_Query:private] => Array ( [0] => init_query_flags [1] => parse_tax_query ) )

Publicaciones relacionadas

Array ( [_edit_lock] => Array ( [0] => 1627900210:1 ) [_edit_last] => Array ( [0] => 1 ) [_wpcf-autor-sort-order] => Array ( [0] => a:1:{i:0;i:22900;} ) [wpcf-autor] => Array ( [0] => César Nebot Monferrer ) [wpcf-director] => Array ( [0] => José García Solanes, Arielle Beyaert Stevens ) [wpcf-fecha-defensa] => Array ( [0] => 1543536000 ) [wpcf-lugar] => Array ( [0] => Universidad de Murcia ) [wpcf-tribunal] => Array ( [0] => Gabriel Pérez Quirós Antonio Montañés Bernal Vicente Orts Ríos ) [wpcf-calificacion] => Array ( [0] => 10 ) )