Triangle Sampling Based Estimation of Average Degree of Network Using Different Estimators

Authors

  • Vivek Kumar Gupta Department of Mathematics and Statistics, Dr. Harisingh Gour Vishwavidyalaya, Sagar, M.P. 470003
  • Diwakar Shukla Department of Mathematics and Statistics, Dr. Harisingh Gour Vishwavidyalaya, Sagar, M.P. 470003

DOI:

https://doi.org/10.31305/rrijm.2022.v07.i05.004

Keywords:

Graph, Sampling, Social network, Triangle, Ratio Estimator, Regression Estimator, Confidence interval(CI), Mean Squared Error(MSE), Bias, Simulation

Abstract

Graphs are used to represent complex relationships which include online social networks, pandemic spread networks and other real world networks. In recent time, graph sampling has been used for the study of different parameters of networks. Such include many sampling algorithms like Random node, Random edge sampling, Rank degree etc. which are used for collection of subsets of a network, but efficient estimation methods are not discussed much for parameter estimation. This paper presents a comparison of different estimators comprises Ratio and Regression estimators to estimate the average degree of a network. Triangle sampling is used to collect sample data using seed nodes. A comparative procedure is used to obtain the lower and upper limits of confidence intervals with the help of multiple triangle samples. Ogive based simulation is also used for single value computation of both limits of confidence interval(CI). The results obtained from the simulation show that Regression estimator is more efficient than the Ratio  in triangle sampling.

Downloads

Published

16-05-2022

How to Cite

Gupta, V. ., & Shukla, D. . (2022). Triangle Sampling Based Estimation of Average Degree of Network Using Different Estimators. RESEARCH REVIEW International Journal of Multidisciplinary, 7(5), 22–36. https://doi.org/10.31305/rrijm.2022.v07.i05.004