BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Vrije Universiteit Amsterdam//NONSGML v1.0//EN NAME:PhD defence G.B. Banava METHOD:PUBLISH BEGIN:VEVENT DTSTART:20260511T134500 DTEND:20260511T151500 DTSTAMP:20260511T134500 UID:2026/phd-defence-g-b-banava@8F96275E-9F55-4B3F-A143-836282E12573 CREATED:20260513T142143 LOCATION:(1st floor) Auditorium, Main building De Boelelaan 1105 1081 HV Amsterdam SUMMARY:PhD defence G.B. Banava X-ALT-DESC;FMTTYPE=text/html:

Targeted Estimation in Heterogeneous Panel Data Models

Predicting more accurately by learning from comparable units

Data scientist Georgia Banava investigated how statistical predictions for specific units, such as a single country or a specific region, can be improved. While standar d econometric models often calculate an average effect across all ava ilable data, Banava developed three new methods that focus specifical ly on one particular unit. This makes it possible, for example, to pr edict the GDP of the Netherlands very precisely, without leaving valu able data from the rest of Europe unused.

Banava's research dem onstrates that 'borrowing' information from comparable units leads to much more reliable, accurate, and stable estimates. Instead of analy zing each unit separately or simply averaging everything, her methods make optimal use of all information while carefully accounting for d ifferences between them. This approach offers major practical benefit s: for instance, hospitals can better evaluate the effectiveness of t reatments per patient group, and smaller regions with limited data ca n still make reliable predictions regarding unemployment or economic growth by learning from comparable regions.

More information on the thesis

DESCRIPTION:

Predicting more accurately by learning from comparabl e units

Data scientist Georgia Banava investigated how statistic al predictions for specific units, such as a single country or a spec ific region, can be improved. While standard econometric models often calculate an average effect across all available data, Banava develo ped three new methods that focus specifically on one particular unit. This makes it possible, for example, to predict the GDP of the Nethe rlands very precisely, without leaving valuable data from the rest of Europe unused. Banava's research demonstrates that 'borrowing' infor mation from comparable units leads to much more reliable, accurate, a nd stable estimates. Instead of analyzing each unit separately or sim ply averaging everything, her methods make optimal use of all informa tion while carefully accounting for differences between them. This ap proach offers major practical benefits: for instance, hospitals can b etter evaluate the effectiveness of treatments per patient group, and smaller regions with limited data can still make reliable prediction s regarding unemployment or economic growth by learning from comparab le regions. More information on the thesis Targeted Estimatio n in Heterogeneous Panel Data Models END:VEVENT END:VCALENDAR