Texas A&M University LibrariesTexas A&M University LibrariesTexas A&M University Libraries
    • Help
    • Login
    OAKTrust
    View Item 
    •   OAKTrust Home
    • Colleges and Schools
    • Graduate and Professional School
    • Electronic Theses, Dissertations, and Records of Study (2002– )
    • View Item
    •   OAKTrust Home
    • Colleges and Schools
    • Graduate and Professional School
    • Electronic Theses, Dissertations, and Records of Study (2002– )
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    METHODOLOGY AND APPLICATIONS IN IMPUTATION, FOOD CONSUMPTION AND OBESITY RESEARCH

    Thumbnail
    View/ Open
    KYUREGHIAN-DISSERTATION.pdf (7.691Mb)
    Date
    2010-07-14
    Author
    Kyureghian, Gayaneh
    Metadata
    Show full item record
    Abstract
    Obesity is a rapidly growing public health threat as well as an economic problem in the United States. The recent changes in eating habits, especially the relative increase of food away from home (FAFH) consumption over the last three decades raised the possibility of causal linkage between obesity and FAFH. This study confirms the positive, significant association between the body mass index and FAFH consumption in adults, consistent with previous findings in the economic and nutrition literature. This work goes a step further, however. We demonstrate FAFH consumption at quick-service restaurants has a significantly larger effect on body mass index than FAFH consumption at full-service restaurants. Further disaggregation of FAFH by meal occasion reveals that lunch consumed away from home has the largest positive effect on body mass index compared to other meal occasions (breakfast, dinner and snacks). Survey data with missing observations or latent variables are not rare phenomena. The missing value imputation methods are combined into two groups, contingent upon the existence or absence of an underlying explicit statistical model. Explicit modeling methods include unconditional mean value imputation, conditional mean and regression imputation, stochastic regression imputation, and multiple imputation. The methods based on implicit modeling include hot deck and cold deck imputation. In the second essay, we review imputation methods commonly used in the agricultural economics literature. Our analysis revealed strong preference of researchers for the regression imputation method. We consider several alternative (regression, mean and median) single imputation methods to impute and to append prices of foods consumed at home (foods commercially purchased and prepared from ingredients) from the National Health and Nutrition Examination Survey (NHANES) dietary intake data. We also demonstrate the superiority of regression imputation method compared to the mean and median imputation methods for commercially prepared foods. For ingredient foods, the results are ambiguous with no imputation method clearly outperforming the others.
    URI
    https://hdl.handle.net/1969.1/ETD-TAMU-2009-05-783
    Subject
    Obesity, Food Away From Home, Nutrition, Imputation, Body Mass Index, Restaurant Type, Meal Occasion
    Collections
    • Electronic Theses, Dissertations, and Records of Study (2002– )
    Citation
    Kyureghian, Gayaneh (2009). METHODOLOGY AND APPLICATIONS IN IMPUTATION, FOOD CONSUMPTION AND OBESITY RESEARCH. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /ETD -TAMU -2009 -05 -783.

    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Advanced Search

    Browse

    All of OAKTrustCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsDepartmentTypeThis CollectionBy Issue DateAuthorsTitlesSubjectsDepartmentType

    My Account

    LoginRegister

    Statistics

    View Usage Statistics
    Help and Documentation

    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV