The Missing Observations in the Time Series of the Dependent Variable and Their Location in the Simple Linear Regression Analysis and their Effect on Prediction (Applied study in textile factory Dhi Qar(

  • Emad Farhood AL-Shareefi Technical College| Dhi_Qar/ Southern Technical University
Keywords: Lost values, observations, simple linear regression, time series

Abstract

             Forecasting, in time series is an important in planning and making assumptions about future events using different statistical methods, and depends on estimating the value of a variable at a future date. The study reviewed the missing views in the time series (a model without loss of observations and three models was assumed to be lost in the views of the dependent variable in different locations in the series) ,After a simple linear regression of the four models of the analysis show that the series without losing it show coherent and clear in their dealings and morally within the statistical acceptable levels, and the loss of view where what is its position within the series and it show obvious effect on the estimated value of any expected value is much greater than the value of truth The Akaike test was used to compare the models and the test results indicated the model's superiority without loss. and has recommended the researcher on the need to use all the views in the dependent variable without loss prediction in the case of a general trend of time and chain succession will researcher estimates far away from the real value which negatively affects the decision-making.

Downloads

Download data is not yet available.
Published
2018-07-31
How to Cite
[1]
E. AL-Shareefi, “The Missing Observations in the Time Series of the Dependent Variable and Their Location in the Simple Linear Regression Analysis and their Effect on Prediction (Applied study in textile factory Dhi Qar(”, JUBH, vol. 26, no. 7, pp. 570 - 579, Jul. 2018.
Section
Articles