DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación
Artículo científico (JournalArticle)
Blind observers of the sky
The concept of blind analysis, a key procedure to remove the human-based systematic error called confirmation bias, has long been an integral part of data analysis in many research areas. In cosmology, blind analysis is recently making its entrance, as the field progresses into a fully fledged high-precision science. The credibility, reliability and robustness of results from future sky-surveys will dramatically increase if the effect of confirmation bias is kept under control by using an appropriate blinding procedure. Here, we present a catalog-level blinding scheme for galaxy clustering data apt to be used in future spectroscopic galaxy surveys. We shift the individual galaxy positions along the line of sight based on 1) a geometric shift mimicking the Alcock-Paczynski effect and 2) a perturbative shift akin to redshift-space distortions. This procedure has several advantages. After combining the two steps above, it is almost impossible to accidentally unblind. The procedure induces a shift in cosmological parameters without changing the galaxies' angular positions, hence without interfering with the effects of angular systematics. Since the method is applied at catalog level, there is no need to adopt several blinding schemes tuned to different summary statistics, likelihood choices or types of analyses. By testing the method on mock catalogs and the BOSS DR12 catalog we demonstrate its performance in blinding galaxy clustering data for relevant cosmological parameters sensitive to the background expansion rate and the growth rate of structures. We publicly release our data products at https://github.com/SamuelBrieden/BlindingCatalogs., Funding for this work was partially provided by the
Spanish MINECO under projects AYA2014-58747-P AEI/FEDER, UE, PGC2018-098866-BI00,FEDER, UE, and MDM-2014-0369 of ICCUB (Unidad de Excelencia Mar´ıa de Maeztu).
HGM and SB acknowledge the support from la Caixa Foundation (ID 100010434) which code
LCF/BQ/PI18/11630024. LV acknowledges support by European Union’s Horizon 2020 research and innovation programme ERC (BePreSySe, grant agreement 725327). JLB is supported by the Allan C. and Dorothy H. Davis Fellowship.
We acknowledge support from the Spanish MICINNs Consolider-Ingenio 2010
Programme under grant MultiDark CSD2009-00064, MINECO Centro de Excelencia Severo
Ochoa Programme under grant SEV- 2012-0249, and grant AYA2014-60641-C2-1-P.
Proyecto:
MINECO / Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016 AYA2014-58747-P
,
AEI / Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020 PGC2018-098866-BI00
,
MINECO / Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016 MDM-2014-0369
,
EC / H2020 725327
,
MINECO / Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016 AYA2014-60641-C2-1-P