Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of British Columbia

$34,000.00

Regime-switching model, finite mixture model, empirical likelihood and other applied problems.

FY 2013-14

Summary
The statistical problems I intend to work on have their applied backgrounds in several areas. One problem is related to the quality of forest products. After breaking 20 pieces of lumber product, we get some idea of how bad the worst quality out of 20 pieces may get, but we are far from certain. How bad the worst may get out of 100 pieces or out of 1000 pieces? Breaking additional pieces will help to answer such questions but the cost will escalate. My statistical suggestion is to combining data from many years and many grades to paint a much clearer quality picture of each year and each grade. Consequently, the per year and per grade costs can be greatly reduced. Removing a tumour together with neighbouring tissues extensively helps to prevent future recurrences. Yet this practice could have unnecessarily caused a greater pain to many patients. Searching for clues in tumour genetic markers is a promising way to predict the probability of recurrence and therefore reduce the number of unwarranted extensive operations. However, doctors must first find that out of millions of markers, which of them are useful predictors. My research helps them to dig deeper into the data they have already collected. With the help of good statistics, they may attain their goal faster and at a lower cost. My research on finite mixture model emphasizes a solid theoretical basis, yet the research problems are motivated by applied problems. In medical applications, my results help them to identify population stratifications. Such findings can be useful in designing personalized treatments. In financial applications, it is vital to classify clients into more homogeneous risk groups. My research provides statistical evidence for the need of stratification and to prevent them from having it over done. My proposed research on HMM is to provide a simpler and more reliable way to analyze financial and other time series. It sheds light on their stochastic nature and is useful at developing most suitable financial products.
Program
Discovery Grants Program - Individual
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
124086-2013
Name nom
Chen, Jiahua
Department département
Statistics
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
2013
Competition year année de concours
2013
Award amount
34000
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
1201
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Life sciences (including biotechnology)
Area of application fr
Sciences de la vie (y compris biotechnologie)
Research subject code
3001
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Statistical theory
Research subject fr
Théorie de statistique
Partie
1
Nb partie
1
Application title
Regime-switching model, finite mixture model, empirical likelihood and other applied problems.
Keyword
NSERC
Application summary
The statistical problems I intend to work on have their applied backgrounds in several areas. One problem is related to the quality of forest products. After breaking 20 pieces of lumber product, we get some idea of how bad the worst quality out of 20 pieces may get, but we are far from certain. How bad the worst may get out of 100 pieces or out of 1000 pieces? Breaking additional pieces will help to answer such questions but the cost will escalate. My statistical suggestion is to combining data from many years and many grades to paint a much clearer quality picture of each year and each grade. Consequently, the per year and per grade costs can be greatly reduced. Removing a tumour together with neighbouring tissues extensively helps to prevent future recurrences. Yet this practice could have unnecessarily caused a greater pain to many patients. Searching for clues in tumour genetic markers is a promising way to predict the probability of recurrence and therefore reduce the number of unwarranted extensive operations. However, doctors must first find that out of millions of markers, which of them are useful predictors. My research helps them to dig deeper into the data they have already collected. With the help of good statistics, they may attain their goal faster and at a lower cost. My research on finite mixture model emphasizes a solid theoretical basis, yet the research problems are motivated by applied problems. In medical applications, my results help them to identify population stratifications. Such findings can be useful in designing personalized treatments. In financial applications, it is vital to classify clients into more homogeneous risk groups. My research provides statistical evidence for the need of stratification and to prevent them from having it over done. My proposed research on HMM is to provide a simpler and more reliable way to analyze financial and other time series. It sheds light on their stochastic nature and is useful at developing most suitable financial products.