A comparison of multiple imputation technique with linear interpolation method for time series data
Loading...
Date
SEPTEMBER, 2019
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis evaluates the performances of Multiple Imputation Technique (MIT)
and Linear Interpolation methods for the estimation of missing values in a time
series data (CO2 emissions data under the Fuel combustion sub-category of the
Energy sector. Under this sub-category, data of two codes namely; i) Energy
industries and ii) Manufacturing Industries and Construction were used).
The performances of both methods were then compared using two notable
indicators; the Mean Absolute Error (MAE) and the Mean-Square Error (RMSE).
This thesis highlights some advantages and limitations of each method compared
with the other, thereby providing suggestions on which method to be used under
prevailing conditions.
Description
A thesis submitted to the Department of Mathematics, Kwame Nkrumah University of Science and Technology, in partial fufillment of the requirement for the degree of Msc. Applied Statistics.
Keywords
MIT, Linear Interpolation methods