# dmt: Samples built upon cosine. In LAJordanger/localgaussSpec: Local Gaussian Spectral Analysis

## Description

The `dmt`-function can create a sample where different trigonometric functions live at separate levels. This function is at the moment rather simplistic, as its main intention is to just create an example for ISNPS-2016 (and I'm running low on time)

## Usage

 ```1 2 3``` ```dmt(n, A, delta, delta_range = NULL, alpha, theta = NULL, wn = list(type = "rnorm", args = list(mean = 0, sd = 1), adjust = 0), seed = NULL) ```

## Arguments

 `n` The number of samples to create. `A` A matrix whose first row should be the positions and second row should be the probabilites of getting there. The cases of main interest to investigate is the case where this describes a process having expectation `0`. `delta` Either a single number, or a vector whose length equals the number of columns in `A`. This will be used as the amplitudes for the cosine-terms. `delta_range` Either `NULL` (the default) or a vector similiar to `delta`. When this is different from `NULL` the effect will be that the values for the deltas will be sampled uniformly based on the intervals that the corresponding components of `delta` and `delta_range` span. `alpha` Either a single number, or a vector whose length equals the number of columns in `A`. This will be used as the frequencies in the cosine-terms. `theta` Either a single number, or a vector whose length equals the number of columns in `A`. This will be used as the phase in the cosine-terms. The default value `NULL` will trigger a procedure that draws these numbers uniformly between `0` and `2pi`. `wn` A list that specifies how to sample from some distribution. The default is to use standard normal white noise. Note: It's assumed that this should be white noise, but there's no test that prevents a user from selecting something different. The value `NULL` can be used to set this term to zero. `seed` An integer to be used for the specification of a seed, the default value NULL will trigger a seed to be selected at random.

## Value

A sample of length `n` based on the specifications given in the other arguments

LAJordanger/localgaussSpec documentation built on Dec. 18, 2018, 2:31 a.m.