Probabilities of large values for sums of i.i.d. non-negative random variables with regular tail of index -1

Abstract: Let

x i_{1}, x i_{2}, d o t s

be i.i.d. non-negative random variables whose tail varies regularly with index

- 1

, let

S_{n}

be the sum and

M_{n}

the largest of the first

n

values. We clarify for which sequences

x_{n} o i n f t y

we have

m a t h b b P (S_{n} g e x_{n}) s i m m a t h b b P (M_{n} g e x_{n})

as

n o i n f t y

. Outside this regime, the typical size of

S_{n}

conditioned on exceeding

x_{n}

is not completely determined by the largest summand and we provide an appropriate correction term which involves the integrated tail of

x i_{1}

.

This page was built for publication: Probabilities of large values for sums of i.i.d. non-negative random variables with regular tail of index $-1$