On the comparison of Fisher information of the Weibull and GE distributions

In this paper, we consider the Fisher information matrices of the generalized exponential (GE) and Weibull distributions for complete and Type-I censored observations. Fisher information matrix can be used to compute asymptotic variances of the different estimators. Although both distributions may provide similar data fit but the corresponding Fisher information matrices can be quite different. Moreover, the percentage loss of information due to truncation of the Weibull distribution is much more than the GE distribution. We compute the total information of the Weibull and GE distributions for different parameter ranges. We compare the asymptotic variances of the median estimators and the average asymptotic variances of all the percentile estimators for complete and Type-I censored observations. One data analysis has been preformed for illustrative purposes. When two fitted distributions are very close to each other and very difficult to discriminate otherwise, the Fisher information or the above mentioned asymptotic variances may be used for discrimination purposes.

[1]  J. Lawless.  Statistical Models and Methods for Lifetime Data , 2002 .

[2]  Asok K. Nanda,et al.  SOME RESULTS ON REVERSED HAZARD RATE ORDERING , 2001 .

[3]  Sangun Park,et al.  Fisher Information in Order Statistics , 1996 .

[4]  D. Kundu,et al.  Theory & Methods: Generalized exponential distributions , 1999 .

[5]  Dilip Roy,et al.  SOME RESULTS ON REVERSED HAZARD RATE , 2001, Probability in the Engineering and Informational Sciences.

[6]  M. J. S. Khan,et al.  A Generalized Exponential Distribution , 1987 .

[7]  Ramesh C. Gupta,et al.  Some Characterization Results Based on Factorization of the (Reversed) Hazard Rate Function , 2004 .

[8]  Debasis Kundu,et al.  Generalized exponential distribution: different method of estimations , 2001 .

[9]  Harshinder Singh,et al.  The Reversed Hazard Rate Function , 1998, Probability in the Engineering and Informational Sciences.

[10]  J. Gastwirth,et al.  WHERE IS THE FISHER INFORMATION IN AN ORDERED SAMPLE , 2000 .

[11]  Jerald F. Lawless,et al.  Statistical Models and Methods for Lifetime Data. , 1983 .

[12]  Debasis Kundu,et al.  Discriminating between the Weibull and log‐normal distributions , 2004 .

[13]  Mohammad Z. Raqab,et al.  Inferences for generalized exponential distribution based on record statistics , 2002 .

[14]  Debasis Kundu,et al.  Discriminating between Weibull and generalized exponential distributions , 2003, Comput. Stat. Data Anal..

[15]  I. Gertsbakh,et al.  On the Fisher information in type-I censored and quantal response data☆ , 1995 .

[16]  Gordon Johnston,et al.  Statistical Models and Methods for Lifetime Data , 2003, Technometrics.

[17]  Gang Zheng.  On the Fisher information matrix in Type II censored data from the exponentiated exponential family , 2002 .

[18]  D. Kundu,et al.  EXPONENTIATED EXPONENTIAL FAMILY: AN ALTERNATIVE TO GAMMA AND WEIBULL DISTRIBUTIONS , 2001 .

[19]  M. Ahsanullah,et al.  Estimation of the location and scale parameters of generalized exponential distribution based on order statistics , 2001 .

[20]  Generalized exponential distribution: Moments of order statistics , 2004 .

[21]  E. L. Lehmann,et al.  Theory of point estimation , 1950 .

[22]  Bradley Efron,et al.  FISHER'S INFORMATION IN TERMS OF THE HAZARD RATE' , 1990 .