Methods For The Analysis Of Explanatory Linear Regression Models With Missing Data At Random
Date
2006-07Author
Nthoiwa, Gothatamang Patrick
Type
ThesisLanguage
enMetadata
Show full item recordAbstract
The analysis of missing data is done with techniques design for the data set that are
completely missing at random as complete case analysis, regression imputation, multiple
imputation and maximum likelihood based method .. however , the data condition required
to minimize the bias derived from an incorrect analysis have not been full determined. In
present work , a simulated data have been carried out to establish the best strategy of
analysis for multiple regression applicable to the data set with values missing at random.
The results shows that the smallest bias is obtained with maximum likelihood based
methods and multiple imputation techniques, although the low percentage of missing
data, and sample size have no effect on the bias of single imputation and complete case
analysis.
Citation
Nthoiwa,G.P.,July,2006.Methods For The Analysis Of Explanatory Linear Regression Models With Missing Data At Random.Publisher
University of Nairobi College of Biological and Physical Sciences