statsmodels.sandbox.tsa.fftarma.ArmaFft.generate_sample

method

ArmaFft.generate_sample(nsample=100, scale=1.0, distrvs=None, axis=0, burnin=0)

Simulate an ARMA

Parameters

nsampleint or tuple of ints

If nsample is an integer, then this creates a 1d timeseries of length size. If nsample is a tuple, creates a len(nsample) dimensional time series where time is indexed along the input variable axis. All series are unless distrvs generates dependent data.

scalefloat

standard deviation of noise

distrvsfunction, random number generator

function that generates the random numbers, and takes sample size as argument default: np.random.randn TODO: change to size argument

burnininteger (default: 0)

to reduce the effect of initial conditions, burnin observations at the beginning of the sample are dropped

axisint

See nsample.

Returns

rvsndarray

random sample(s) of arma process

Notes

Should work for n-dimensional with time series along axis, but not tested yet. Processes are sampled independently.