BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ZContent.net//ZapCalLib 1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTART;TZID=Atlantic/Canary:20140923T123000
DTEND;TZID=Atlantic/Canary:20140923T133000
UID:iactalks-670
X-WR-CALNAME: IAC Talks: Open Astronomy Seminars
X-ORIGINAL-URL: /iactalks/Talks/view/670
CREATED:2014-09-23T12:30:00+01:00
X-WR-CALDESC: IAC Talks upcomming talks
SUMMARY:Microvariability detection in quasars: An astrostatistical problem
DESCRIPTION:Microvariability detection in quasars: An astrostatistical prob
 lem\nDr. José A.  de Diego Onsurbe\n\nMicrovariations probe the physics a
 nd internal structure of quasars. Unpredictability and small flux variatio
 ns make this phenomenon elusive and difficult to detect. Variance based pr
 obes such as the C and F tests, or a combination of both, are popular meth
 ods to compare the light-curves of the quasar and a comparison star. Recen
 tly, detection claims in some studies depend on the agreement of the resul
 ts of the C and F tests, or of two instances of the F-test, in rejecting t
 he non-variation null hypothesis. However, the C-test is a non-reliable st
 atistical procedure, the F-test is not robust, and the combination of test
 s with concurrent results is anything but a straightforward methodology. A
  priori Power Analysis calculations and post hoc analysis of Monte-Carlo s
 imulations show excellent agreement for the Analysis of Variance test to d
 etect microvariations, as well as the limitations of the F-test. Additiona
 lly, combined tests yield correlated probabilities that make the assessmen
 t of statistical significance unworkable. However, it is possible to inclu
 de data from several field stars to enhance the power in a single F - test
  or ANOVA nested designs, increasing the reliability of the statistical an
 alysis. These would be the preferred methodology when several comparison s
 tars are available. These results show the importance of using adequate me
 thodologies, and avoid inappropriate procedures that can jeopardize microv
 ariability detections. Power analysis and Monte-Carlo simulations are usef
 ul tools for research planning, as they can reveal the robustness and reli
 ability of different research approaches.
END:VEVENT
END:VCALENDAR