BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
 M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20261019T103644Z
DTSTART:20261021T130000Z
DTEND:20261021T140000Z
SUMMARY:Statistics Seminar - Hera Shi (UCL)
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}dd3-mvf45wd
 x-cf0lc9
DESCRIPTION:Title: Empirical Bayes Transfer Learning for Cross-Ancestry M
 endelian Randomization\n\nAbstract: Limited genome-wide association stud
 y (GWAS) sample sizes in underrepresented populations can reduce the pre
 cision of Mendelian randomization (MR) through uncertainty in instrument
  strengths. Associations from other ancestries may improve precision\, b
 ut borrowing must accommodate population-specific causal effects and ple
 iotropy. Motivated by this problem\, we develop \\emph{MR-NPEB}\, a nonp
 arametric empirical Bayes method for heteroscedastic multivariate errors
 -in-variables models with general loading matrices. MR-NPEB estimates th
 e joint distribution of instrument strengths and incorporates their post
 erior predictions into a one-step score estimator of causal effects and 
 pleiotropy variances. In cross-ancestry MR\, this enables borrowing thro
 ugh the dependence of genetic effects without imposing common causal par
 ameters. Under a many-weak-instrument regime\, we establish identificati
 on\, asymptotic normality\, and consistent covariance estimation\, with 
 no first-order variance cost from posterior estimation under suitable co
 nditions. Our efficiency analysis distinguishes two gains: learning the 
 effect distribution yields no larger asymptotic covariance than MR using
  robust adjusted profile scores (MR-RAPS)\, while auxiliary associations
  further improve precision through posterior prediction of target instru
 ment strengths. We characterize the borrowing gain in terms of cross-pop
 ulation dependence and sampling precision\, and establish semiparametric
  efficiency under homoscedasticity and completeness. Simulations and an 
 analysis of Mexican Biobank GWAS data using UK Biobank associations illu
 strate the improvements in causal-effect precision.\n
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
LOCATION:Frank Adams 2\, Alan Turing Building\, Manchester
END:VEVENT
END:VCALENDAR