Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

On Scale and Exponent Functions of a Class of Limiting Processes


Affiliations
1 Department of Statistics, University of Mysore, Manasagangotri, Mysore-570 006, India
     

   Subscribe/Renew Journal


Suppose {Xn, n ∈ Z+) - see below for the notations used - is a sequence of random variables (rv's) such that, for every t > 0 and for ψ given by (1.1) below,

ψ(X(nt)nn)->Y(t),

where Y\l) (in particular) is (assumed) nondegenerate. Then it is shown that, for some positive functions α and β on (0,∞) and for every t > 0,

Y(t)=ψ(Y(1);α(t),β(t))

Explicit expressions for a and p are derived. As an application, the (Pancheva) class of all possible limit laws for power normalized partial maxima of i.i.d rv's ("the p-max stable laws") is re-derived.


Keywords

Scale and Exponential Functions, P-Max Stable Laws, Limiting Processes.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 177

PDF Views: 0




  • On Scale and Exponent Functions of a Class of Limiting Processes

Abstract Views: 177  |  PDF Views: 0

Authors

N. R. Mohan
Department of Statistics, University of Mysore, Manasagangotri, Mysore-570 006, India
S. Ravi
Department of Statistics, University of Mysore, Manasagangotri, Mysore-570 006, India

Abstract


Suppose {Xn, n ∈ Z+) - see below for the notations used - is a sequence of random variables (rv's) such that, for every t > 0 and for ψ given by (1.1) below,

ψ(X(nt)nn)->Y(t),

where Y\l) (in particular) is (assumed) nondegenerate. Then it is shown that, for some positive functions α and β on (0,∞) and for every t > 0,

Y(t)=ψ(Y(1);α(t),β(t))

Explicit expressions for a and p are derived. As an application, the (Pancheva) class of all possible limit laws for power normalized partial maxima of i.i.d rv's ("the p-max stable laws") is re-derived.


Keywords


Scale and Exponential Functions, P-Max Stable Laws, Limiting Processes.