Simulate from Multivariate Normal Density for Metropolis Algorithm

Provided for Random Walk Metropolis algorithm

qprop(theta1, nu)

Arguments

theta1	Vector of current quantiles
nu	Either a single numeric value for the covariance matrix, or a vector for the diagonal

Value

Returns a single numeric simulated value from a Normal distribution or vector of length theta1. length(mu) matrix with one sample in each row.

References

B. D. Ripley (1987) Stochastic Simulation. Wiley. Page 98

Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth edition. Springer.

Examples

s <- replicate(1000, qprop(0, 1))
summary(s)
#>     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
#> -3.28849 -0.67223  0.07392  0.04726  0.74391  4.02702 
hist(s, col='light blue')