Comparing Edgeworth Expansion and Saddlepoint Approximation in Assessing the Asymptotic Normality Behavior of A Non-Parametric Estimator for Finite Population Total



Abstract Views
171

Downloads
158

Citations

Share

##plugins.themes.bootstrap3.article.sidebar##

Submitted Nov 1, 2022
Published Jan 31, 2023
10.24018/ejmath.2023.4.1.167

Abstract viewed = 171 times

Table of Contents

##plugins.themes.bootstrap3.article.main##

Sample surveys concern themselves with drawing inferences about the population based on sample statistics. We assess the asymptotic normality behavior of a proposed nonparametric estimator for finite a population total based on Edgeworth expansion and Saddlepoint approximation. Three properties; unbiasedness, efficiency and coverage probability of the proposed estimators are compared. Based on the background of the two techniques, we focus on confidence interval and coverage probabilities. Simulations on three theoretical data variables in R, revealed that Saddlepoint approximation performed better than Edgeworth expansion. Saddlepoint approximation resulted into a smaller MSE, tighter confidence interval length and higher coverage probability compared to Edgeworth Expansion. The two techniques should be improved in estimation of parameters in other sampling schemes like cluster sampling.
Keywords: Edgeworth expansion Nonparametric estimator auxiliary variables Saddlepoint approximation

References

  1. Odhiambo R, Mwalili S. Nonparametric regression for finite population estimation. East African Journal of Statistics. 2000; (Part 2): 107-118.
     Google Scholar
  2. Laszlo GAKM, Walk H. A Distribution Free-Theory of Nonparametric Regression. Springer-Verlag, New York, 2002.
     Google Scholar
  3. Dorfman AH, Hall P. Estimators of the finite population distribution function using nonparametric regression. Journal of the American Statistical Association. 1992; 21: 1452-1475.
     Google Scholar
  4. Chambers JM. On Methods of Asymptotic Approximation for Multivariate Distributions. Biometrica. 1967.
     Google Scholar
  5. Dorfman AH. Nonparametric regression for estimating totals in finite population. Journal of the American Statistical Association. 1992; 4: 622-625.
     Google Scholar
  6. Breidt F, Opsomer P. Model-Assisted estimator for Complex Surveys using Penalized Splines. Biometrica. 2005; 92(4).
     Google Scholar
  7. Ombui T. Robust Estimation of Finite Population Total Using Local Polynomial Regression. PhD Thesis, Jomo Kenyatta University of Agriculture and Technology, 2008.
     Google Scholar
  8. Daniels HE. Saddlepoint approximations. Annals of Mathematical Statistics. 1987; 25: 631-650.
     Google Scholar
  9. Lugannani R, Rice S. Saddle point approximation for the distribution of the sum of independent random variables. Advances in Applied Probability. 1980; 12: 475-490.
     Google Scholar
  10. Easton GS. General Saddlepoint approximation with applications to statistics. Journal of the American Statistical Association. 2008; 81: 420-430.
     Google Scholar
  11. Okungu JO, Orwa GO, Odhiambo RO. Nonparametric Estimator of a Finite Population Total Based on Saddle Point Approximation. International Journal of Statistics and Applications. 2020; 10(3): 60-67.
     Google Scholar
  12. Okungu JO, Orwa GO, Odhiambo RO. Nonparametric Estimator of a Finite Population Total Based on Edgeworth Expansion. Science Journal of Applied Mathematics and Statistics. 2020; 8(2): 35-41.
     Google Scholar