Grant: Discovery Grants Program - Individual
University of British Columbia
$14,991.00
New Approaches in the Design of Experiments
FY 2012-13
Summary
In many physical experiments, the goal is to purposefully change the settings of some set of factors and measure their effects on a response. Typically, these factors are varied together so that interactions between these factors can be measured. Setting these factors and collecting a measurement is referred to as a run. Often the number of runs is limited, and the experimenter must sacrifice estimating some of these effects.
The projects in this proposal cover two separate areas in the field of design of experiments. The first area in this project deals with the case where prior knowledge suggests a certain set of interactions may be important, and it is necessary to find a design to estimate these effects in the number of runs available. Since the estimated effects can be biased by those that are not estimated, this potential bias is also taken into account.
The second area of research is concerned with finding designs that make the settings of the runs as different as possible. Such problems arise in providing inputs to computer simulations that model physical processes.
One way of measuring this difference between runs is to ensure that the runs are different when smaller subsets of factors are considered. This research will investigate how to measure the quality of a design, and methods for constructing designs.
Program
Discovery Grants Program - Individual
Recipient
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
418422-2012
Name nom
Lekivetz, Ryan
Department département
Okanagan - Mathematics, Statistics and Physics Unit
Organization id
2
Institution établissement
University of British Columbia
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2012
Competition year année de concours
2012
Award amount
14991
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
9999
Area of application group en
Not available
Area of application group fr
Non disponible
Area of application en
Not available
Area of application fr
Non disponible
Research subject code
3000
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Statistics and probability
Research subject fr
Statistique et probabilité
Partie
1
Nb partie
1
Application title
New Approaches in the Design of Experiments
Keyword
NSERC
Application summary
In many physical experiments, the goal is to purposefully change the settings of some set of factors and measure their effects on a response. Typically, these factors are varied together so that interactions between these factors can be measured. Setting these factors and collecting a measurement is referred to as a run. Often the number of runs is limited, and the experimenter must sacrifice estimating some of these effects.
The projects in this proposal cover two separate areas in the field of design of experiments. The first area in this project deals with the case where prior knowledge suggests a certain set of interactions may be important, and it is necessary to find a design to estimate these effects in the number of runs available. Since the estimated effects can be biased by those that are not estimated, this potential bias is also taken into account.
The second area of research is concerned with finding designs that make the settings of the runs as different as possible. Such problems arise in providing inputs to computer simulations that model physical processes.
One way of measuring this difference between runs is to ensure that the runs are different when smaller subsets of factors are considered. This research will investigate how to measure the quality of a design, and methods for constructing designs.