Testing Procedure for Scale Shift at an Unknown Time Point

  • Song, Il-Seong (Professor, Department of Statistics, Sungshim Women's University, Seoul, 136-742)
  • Published : 1996.04.01

Abstract

A testing procedure is considered to the problem of testing whether there exists a shift in scale at an unknown time point whem a fixed number of observations are drawn successively in time. A test statistic based on squared ranks test for equal variances is suggested and its aymptotic distrbution is dereived. Small sample power comparisons are performed.

Keywords

References

  1. Journal of the American Statistical Association v.77 The identical distribution hypothesis for stock market prices: location- and scale-shift alternatives Ali, M.M.;Giaccotto, C.
  2. Technometrics v.28 An efficiency robust nonparametric test for scale change for data from a gamma distribution Gastwirth, J.L.;Mahmoud, H.
  3. Theory of Rank Tests Hajek, J.;Sidak, Z.
  4. Communications in Statistics-Theory and Methods v.13 Nonparametric tests for scale shift at an unknown time point Hsieh, H.K.
  5. Applied Statistics v.26 Tests for variance shift at an unknown time point Hsu, D.A.
  6. Journal of the American Statistical Association v.74 Detecting shifts of parameter in gamma sequences with applications to stock price and air traffic flow analysis Hsu, D.A.
  7. Applied Statistics v.28 A non-parametric approach to the change-point problem Pettitt, A.N.
  8. Applied Statistics v.30 Detecting a scale shift in a random sequence at an unknown time point Talwar, P.P.;Gentle, J.E.