• Title/Summary/Keyword: statistics

Search Result 23,275, Processing Time 0.048 seconds

Categorical Data Analysis by Means of Echelon Analysis with Spatial Scan Statistics

  • Moon, Sung-Ho
    • Journal of the Korean Data and Information Science Society
    • /
    • v.15 no.1
    • /
    • pp.83-94
    • /
    • 2004
  • In this study we analyze categorical data by means of spatial statistics and echelon analysis. To do this, we first determine the hierarchical structure of a given contingency table by using echelon dendrogram then, we detect candidates of hotspots given as the top echelon in the dendrogram. Next, we evaluate spatial scan statistics for the zones of significantly high or low rates based on the likelihood ratio. Finally, we detect hotspots of any size and shape based on spatial scan statistics.

  • PDF

Recurrence Relations Between Product Moments of Order Statistics for Truncated Distributions and Their Applications

  • Saran, Jagdish;Pushkarna, Narinder
    • Journal of the Korean Statistical Society
    • /
    • v.31 no.3
    • /
    • pp.391-403
    • /
    • 2002
  • In this paper, some general results for obtaining recurrence relations between product moments of order statistics for doubly truncated distributions are established. These results are then applied to some specific doubly truncated distributions, viz. doubly truncated Weibull, Exponential, Pareto, power function, Cauchy, Lomax and Rayleigh.

CHARACTERIZATION OF CONTINUOUS DISTRIBUTIONS THROUGH RECORD STATISTICS

  • Khan, Abdul Hamid;Faizan, Mohd.;Haque, Ziaul
    • Communications of the Korean Mathematical Society
    • /
    • v.25 no.3
    • /
    • pp.485-489
    • /
    • 2010
  • A family of continuous probability distribution has been characterized through the difference of two conditional expectations, conditioned on a non-adjacent record statistic. Also, a result based on the unconditional expectation and a conditional expectation is used to characterize a family of distributions. Further, some of its deductions are also discussed.

A Bayesian Approach for Record Value Statistics Model Using Nonhomogeneous Poisson Process

  • Kiheon Choi;Hee chual Kim
    • Communications for Statistical Applications and Methods
    • /
    • v.4 no.1
    • /
    • pp.259-269
    • /
    • 1997
  • Bayesian inference for a record value statistics(RVS) model of nonhomogeneous Poisson process is considered. We seal with Bayesian inference for double exponential, Gamma, Rayleigh, Gumble RVS models using Gibbs sampling and Metropolis algorithm and also explore Bayesian computation and model selection.

  • PDF

Recurrence Relations in the Transformed Exponential Distributions

  • Choi, Jeen-Kap;Mo, Kap-Jong
    • Journal of the Korean Data and Information Science Society
    • /
    • v.14 no.4
    • /
    • pp.1031-1044
    • /
    • 2003
  • In this paper, we establish some recurrence relations of the moments, product moments, percentage points, and modes of order statistics from the transformed exponential distribution.

  • PDF

Utilizing Order Statistics in Density Estimation

  • Kim, W.C.;Park, B.U.
    • Communications for Statistical Applications and Methods
    • /
    • v.2 no.2
    • /
    • pp.227-230
    • /
    • 1995
  • In this paper, we discuss simple ways of implementing non-basic kernel density estimators which typically ceed extra pilot estimation. The methods utilize order statistics at the pilot estimation stages. We focus mainly on bariable lacation and scale kernel density estimator (Jones, Hu and McKay, 1994), but the same idea can be applied to other methods too.

  • PDF