def make_trend(series_len, rw_loc=0.001, rw_scale=0.1, type='rw', seed=1):
""" Module to generate time-series trend with different methods
Parameters
----------
series_len: int
total length of series
type: str ['arma', 'rw']
rw_loc: float
mean of random walk
rw_scale: float
scale of random walk
seed: int
Returns
-------
np.array-llike with length equals `series_len`
"""
# make trend
if type == "rw":
rw = np.random.default_rng(seed).normal(rw_loc, rw_scale, series_len)
trend = np.cumsum(rw)
elif type == "arma":
# TODO: consider parameterize this
arparams = np.array([.25])
maparams = np.array([.6])
ar = np.r_[1, -arparams]
ma = np.r_[1, maparams]
arma_process = sm.tsa.ArmaProcess(ar, ma)
trend = arma_process.generate_sample(series_len)
else:
raise IllegalArgument("Invalid trend_type.")
return trend
def make_trend(series_len,
method='rw',
arma=[.25, .6],
rw_loc=0.0,
rw_scale=0.1,
seed=1):
""" Module to generate time-series trend with different methods
Parameters
----------
series_len: int
Total length of series
method: str ['arma', 'rw']
In case of `'rw'`, a simple random walk process will be used. For `'arma'`, we will use `statsmodels.api` to
simulate a simple ARMA(1, 1) process
arma: list
List [arparams, maparams] of size 2 where used for arma(1) generating process
rw_loc: float
Location parameter of random walk generated by `np.random.normal()`
rw_scale: float
Scale parameter of random walk generated by `np.random.normal()`
seed: int
Seed passed into `np.random.default_rng()`
Returns
-------
np.array-llike
Simulated trend with length equals `series_len`
Notes
-----
1. ARMA process: https://www.statsmodels.org/stable/generated/statsmodels.tsa.arima_process.ArmaProcess.html
"""
# make trend
if method == "rw":
rw = np.random.default_rng(seed).normal(rw_loc, rw_scale, series_len)
trend = np.cumsum(rw)
elif method == "arma":
arparams = np.array([arma[0]])
maparams = np.array([arma[1]])
# add zero-lag and negate
ar = np.r_[1, -arparams]
# add zero-lag
ma = np.r_[1, maparams]
arma_process = ArmaProcess(ar, ma)
trend = arma_process.generate_sample(series_len)
else:
raise IllegalArgument("Invalid trend method.")
return trend
def make_seasonal_dummies(df, date_col, freq, sparse=True, drop_first=True):
""" Based on the frequency input (in pandas.DataFrame style), provide dummies indicator for regression type of
purpose.
Parameters
----------
df : pd.DataFrame
Input dataframe to supply datetime array for generating series of indicators
date_col : str
Label of the date column supply for generating series
freq : str ['weekday', 'month', 'week']
Options to pick the right frequency for generating dummies
sparse : bool
argument passed into `pd.get_dummies`
drop_first : bool
argument passed into `pd.get_dummies`
Returns
-------
out : pd.DataFrame
data with computed fourier series attached
fs_cols : list
list of labels derived from fourier series
Notes
-----
This is calling :func:`pd.get_dummies`
"""
if freq == 'weekday':
dummies = pd.get_dummies(df[date_col].dt.weekday,
prefix='wd',
sparse=sparse,
drop_first=drop_first)
elif freq == 'month':
dummies = pd.get_dummies(df[date_col].dt.month,
prefix='m',
sparse=sparse,
drop_first=drop_first)
elif freq == 'week':
dummies = pd.get_dummies(df[date_col].dt.week,
prefix='w',
sparse=sparse,
drop_first=drop_first)
else:
raise IllegalArgument("Invalid argument of freq.")
cols = dummies.columns.tolist()
out = pd.concat([df, dummies], axis=1)
return out, cols