Announcement: Nixtla Enterprise now offers top foundation models, MCP, and agentic capabilities: join the waitlist
statsforecast.etsetssimulateetssimulate(
x: ndarray,
m: int,
error: Component,
trend: Component,
season: Component,
alpha: float,
beta: float,
gamma: float,
phi: float,
h: int,
y: ndarray,
e: ndarray
) → None
etsforecastetsforecast(
x: ndarray,
m: int,
trend: Component,
season: Component,
phi: float,
h: int,
f: ndarray
) → None
initparaminitparam(
alpha: float,
beta: float,
gamma: float,
phi: float,
trendtype: str,
seasontype: str,
damped: bool,
lower: ndarray,
upper: ndarray,
m: int,
bounds: str
)
admissibleadmissible(alpha: float, beta: float, gamma: float, phi: float, m: int)
check_paramcheck_param(
alpha: float,
beta: float,
gamma: float,
phi: float,
lower: ndarray,
upper: ndarray,
bounds: str,
m: int
)
fourierfourier(x, period, K, h=None)
initstateinitstate(y, m, trendtype, seasontype)
switchswitch(x: str) → Component
switch_criterionswitch_criterion(x: str) → Criterion
pegelsresid_Cpegelsresid_C(
y: ndarray,
m: int,
init_state: ndarray,
errortype: str,
trendtype: str,
seasontype: str,
damped: bool,
alpha: float,
beta: float,
gamma: float,
phi: float,
nmse: int
)
optimize_ets_target_fnoptimize_ets_target_fn(
x0,
par,
y,
nstate,
errortype,
trendtype,
seasontype,
damped,
par_noopt,
lowerb,
upperb,
opt_crit,
nmse,
bounds,
m,
pnames,
pnames2
)
etsmodeletsmodel(
y: ndarray,
m: int,
errortype: str,
trendtype: str,
seasontype: str,
damped: bool,
alpha: float,
beta: float,
gamma: float,
phi: float,
lower: ndarray,
upper: ndarray,
opt_crit: str,
nmse: int,
bounds: str,
maxit: int = 2000,
control=None,
seed=None,
trace: bool = False
)
is_constantis_constant(x)
ets_fets_f(
y,
m,
model='ZZZ',
damped=None,
alpha=None,
beta=None,
gamma=None,
phi=None,
additive_only=None,
blambda=None,
biasadj=None,
lower=None,
upper=None,
opt_crit='lik',
nmse=3,
bounds='both',
ic='aicc',
restrict=True,
allow_multiplicative_trend=False,
use_initial_values=False,
maxit=2000
)
pegelsfcast_Cpegelsfcast_C(h, obj, npaths=None, level=None, bootstrap=None)
forecast_etsforecast_ets(obj, h, level=None)
forward_etsforward_ets(fitted_model, y)
Was this page helpful?