How can I introduce a known autocovariance to a lognormal process

Hi,

I am trying to do a montecarlo simulation for a lognormal process with known mean, variance and autocovariance

I have used logrnd to generate this process as:

y = lognrnd(mu,sigma,[1,simpara]); % where simpara is the simulation parameter so the length of the vector

now I want to introduce the autocovariance to y. I have tried the multiplication with covariance matrix but it seems that it doesn't work.

I am thinking about using the command filter but not sure how to set the nominator and donominator of the filter.

Could you please help me?

Thank you so much in advance

Answer 1

[Jeff Miller]

You should be able to do it by generating your sequence one element at a time, if you can't find a canned routine that will do it automagically. The first element can be generated randomly, but then the second element has to be generated conditionally on the value of the first one, the third has to be generated conditionally on the value of the second one, and so on. I'm not familiar with autocorrelated lognormal processes, but the definition of the autocovariance for this process should tell you the distribution of the element at position n+1 conditional on the value of the element at position n.