earticle

논문검색

Effects of Missing Value Estimation Methods in Correlation Matrix-A Case Study of Concrete Compressive Strength Data

초록

영어

Concrete compressive strength is one of the most important factors leading to building construction, in the civil engineering context. While evaluating such data, quantitative analysis required. As it is known that, concrete as a non-homogeneous material, consists of separate phases. The more complicated the concrete, the higher is the compressive strength. But if missing value exists in the microstructure of concrete, then it may provide some unusual effect on the compressive strength of concrete. Thus it is required to deal with the analysis of missing values. In this study traditional and modern estimation techniques of missing values are performed and the effect of these methods on correlation matrix is observed along with their comparison. The result shows that, modern techniques provide efficient estimates compared to traditional method. .The analysis described here were undertaken in the SPSS 13.0 packages.

목차

Abstract
 1. Introduction
 2. Missing Values
  2.1 Missing Completely At Random (MCAR)
  2.2 Missing At Random (MAR)
  2.3 Non Ignorable (NI) Missing Values
 3. Traditional Approaches of Missing Values
  3.1 Deletion Techniques
  3.2 Imputation Techniques
  3.3 Modern Alternatives For Working With Missing Values
 4. Results of Empirical Application
 5. Conclusion
 References

저자정보

  • Azizur Rahman Lecturer, Department of Statistics, Jagannath University
  • Ajit Kumar Majumder Professor, Department of Statistics, Jahangirnagar University, Savar, Dhaka, Bangladesh

참고문헌

    ※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.

    0개의 논문이 장바구니에 담겼습니다.