wnRT {weakARMA}

Weak white noise

Description

Simulates an uncorrelated but dependant noise process.

Usage

wnRT(n, sigma = 1, k = 1, ninit = 100)

Arguments

n

Number of observations.

sigma

Standard deviation.

k

Integer \neq 0 to prevent a zero denominator.

ninit

Length of 'burn-in' period.

Value

Vector of size n containing a nonlinear sequence X_{i} such as X_i = \frac{Z_{i}}{|Z_{i+1}| + k} , where Z_{i} is a sequence of iid random variables mean-zero random variable with variance \sigma^2.

References

Romano, J. and Thombs, L. 1996, Inference for autocorrelation under weak assumptions, Journal of the American Statistical Association, vol. 91, no. 434, pp. 590-600

See Also

wnPT, wnPT_SQ, simGARCH

Examples

wnRT(100)
wnRT(100, sigma = 1)