Package 'arima2'

ARIMA AIC table

Description

Construct table of AIC for all combinations 0<=p<=P and 0<=q<=Q

Usage

aicTable(data, P, Q, D = 0, ic = c("aic", "aicc"), ...)

Arguments

data	a time series object, or a dataset that can be used as input into the arima function.
P	a positive integer value representing the maximum number of AR coefficients that should be included in the table.
Q	a positive integer value representing the maximum number of MA coefficients that should be included in the table.
D	a positive integer value representing the degree of differencing
ic	Information criterion to be used in the table.
...	Additional arguments passed to arima().

Details

This function creates an AIC table for ARMA models of varying sizes. Each row for the table corresponds to a different AR value, and each column of the table corresponds to a different MA value.

Value

A matrix containing the model AIC values.

Examples

set.seed(654321)
aicTable(presidents, 3, 2)

---

ARIMA Modeling of Time Series

Description

Fit an ARIMA model to a univariate time series. This function builds on the ARIMA model fitting approach used in stats::arima() by fitting model parameters via a random restart algorithm.

Usage

arima(
  x,
  order = c(0L, 0L, 0L),
  seasonal = list(order = c(0L, 0L, 0L), period = NA),
  xreg = NULL,
  include.mean = TRUE,
  transform.pars = TRUE,
  fixed = NULL,
  init = NULL,
  method = c("CSS-ML", "ML", "CSS"),
  n.cond,
  SSinit = c("Rossignol2011", "Gardner1980"),
  optim.method = "BFGS",
  optim.control = list(),
  kappa = 1e+06,
  diffuseControl = TRUE,
  max_iters = 100,
  max_repeats = 10,
  max_inv_root = 1,
  min_inv_root_dist = 0,
  eps_tol = 1e-04
)

Arguments

x	a univariate time series
order	A specification of the non-seasonal part of the ARIMA model: the three integer components $(p, d, q)$ are the AR order, the degree of differencing, and the MA order.
seasonal	A specification of the seasonal part of the ARIMA model, plus the period (which defaults to frequency(x)). This may be a list with components order and period, or just a numeric vector of length 3 which specifies the seasonal order. In the latter case the default period is used.
xreg	Optionally, a vector or matrix of external regressors, which must have the same number of rows as x.
include.mean	Should the ARMA model include a mean/intercept term? The default is TRUE for undifferenced series, and it is ignored for ARIMA models with differencing.
transform.pars	logical; if true, the AR parameters are transformed to ensure that they remain in the region of stationarity. Not used for method = "CSS". For method = "ML", it has been advantageous to set transform.pars = FALSE in some cases, see also fixed.
fixed	optional numeric vector of the same length as the total number of coefficients to be estimated. It should be of the form $(\phi_1, \ldots, \phi_p, \theta_1, \ldots, \theta_q, \Phi_1, \ldots, \Phi_P, \Theta_1, \ldots, \Theta_Q, \mu),$ where $\phi_i$ are the AR coefficients, $\theta_i$ are the MA coefficients, $\Phi_i$ are the seasonal AR coefficients, $\Theta_i$ are the seasonal MA coefficients and $\mu$ is the intercept term. Note that the $\mu$ entry is required if and only if include.mean is TRUE. In particular it should not be present if the model is an ARIMA model with differencing. The entries of the fixed vector should consist of the values at which the user wishes to “fix” the corresponding coefficient, or NA if that coefficient should not be fixed, but estimated. The argument transform.pars will be set to FALSE if any AR parameters are fixed. A warning will be given if transform.pars is set to (or left at its default) TRUE. It may be wise to set transform.pars = FALSE even when fixing MA parameters, especially at values that cause the model to be nearly non-invertible.
init	optional numeric vector of initial parameter values. Missing values will be filled in, by zeroes except for regression coefficients. Values already specified in fixed will be ignored.
method	fitting method: maximum likelihood or minimize conditional sum-of-squares. The default (unless there are missing values) is to use conditional-sum-of-squares to find starting values, then maximum likelihood. Can be abbreviated.
n.cond	only used if fitting by conditional-sum-of-squares: the number of initial observations to ignore. It will be ignored if less than the maximum lag of an AR term.
SSinit	a string specifying the algorithm to compute the state-space initialization of the likelihood; see KalmanLike for details. Can be abbreviated.
optim.method	The value passed as the method argument to optim.
optim.control	List of control parameters for optim.
kappa	the prior variance (as a multiple of the innovations variance) for the past observations in a differenced model. Do not reduce this.
diffuseControl	Boolean indicator of whether or initial observations will have likelihood values ignored if controlled by the diffuse prior, i.e., have a Kalman gain of at least 1e4.
max_iters	Maximum number of random restarts for methods "CSS-ML" and "ML". If set to 1, the results of this algorithm is the same as stats::arima() if argument diffuseControl is also set as TRUE. max_iters is often not reached because the condition max_repeats is typically achieved first.
max_repeats	Integer. If the last max_repeats random starts did not result in improved likelihoods, then stop the search. Each result of the optim function is only considered to improve the likelihood if it does so by more than eps_tol.
max_inv_root	positive numeric value less than or equal to 1. This number represents the maximum size of the inverted MA or AR polynomial roots for a new parameter estimate to be considered an improvement to previous estimates. Concerns of numeric stability arise when the size of polynomial roots are near unity circle. The default value 1 means that the the parameter values corresponding with the best log-likelihood will be returned, even if they are near unity. Suitable values of this parameter are near the value 1.
min_inv_root_dist	positive numeric value less than 1. This number represents the minimum distance between AR and MA polynomial roots for a new parameter estimate to be considered an improvement on previous estimates. This is intended to avoid the possibility of returning parameter estimates with nearly canceling roots. Appropriate choices are values near 0.
eps_tol	Tolerance for accepting a new solution to be better than a previous solution in terms of log-likelihood. The default corresponds to a one ten-thousandth unit increase in log-likelihood.

Value

A list of class c("Arima2", "Arima"). This list contains all of the same elements as the output of stats::arima, along with some additional elements. All elements of the output list are:

coef

A vector of AR, MA, and regression coefficients. These can be extracted by the stats::coef method.

sigma2

The MLE of the variance of the innovations.

var.coef

The estimated variance matrix of the coefficients coef, which can be extracted by the stats::vcov method.

mask

A vector containing boolean values, indicating which parameters of the model were estimated.

loglik

The maximized log-likelihood (of the differenced data).

aic

The AIC value corresponding to the log-likelihood.

arma

A compact form of the model specification, as a vector giving the number of AR, MA, seasonal AR and seasonal MA coefficients, plus the period and the number of non-seasonal and seasonal differences.

residuals

The fitted innovations.

call

The matched call.

series

The name of the series x.

code

The convergence value returned by stats::optim.

n.cond

The number of initial observations not used in the fitting.

nobs

The number of observations used for the fitting.

model

A list representing the Kalman Filter used in the fitting.

x

The input time series.

num_starts

Number of restarts before convergence criteria was satisfied.

all_values

Numeric vector of length num_starts containing the loglikelihood of every parameter initialization.

Examples

# example code
set.seed(12345)
arima(miHuron_level$Average, order = c(2, 0, 1), max_iters = 100)

---

ARMA polyroots

Description

This function calculates the roots of the AR or MA polynomials that correspond to an ARMA model.

Usage

ARMApolyroots(model, type = c("AR", "MA"))

Arguments

model	Either of fitted object of class Arima (i.e., the output of either stats::arima() or arima), a list with named elements at least one of the named elements ar or ma, or a vector with named elements, such as c("ar1" = 0.3, "ar2" = -0.2, "ma1" = 0.14) Seasonal coefficients are ignored.
type	character of value "AR" or "MA", indicating whether or not the AR or MA polynomial roots are desired.

Value

A numeric vector containing the roots of the MA or AR polynomials

Examples

set.seed(123456)
ARMApolyroots(sample_ARMA_coef((order = c(2, 1))), type = "AR")

mod <- arima(lh, order = c(3,0,0))
ARMApolyroots(mod, type = "AR")

---

Annual January levels of lake Michigan-Huron

Description

The dataset is a subset of the monthly average depth (ft) of lake Michigan-Huron. The data were retrieved online from the Great Lakes Environmental Research Laboratory. Various measurement gauges exist; this data was taken from the master gauge.

Usage

miHuron_level

Format

miHuron_level

A data frame with 155 observations and two columns:

Date

Date column that records when the observation was made.

Average

numeric column representing the average depth in feet.

Source

https://www.glerl.noaa.gov/data/dashboard/data/levels/mGauge/miHuronMog.csv

---

Plot Arima2 object

Description

This function plots time series data loaded from an Arima2 object or plots inverse roots of the AR or MA polynomials in a fitted ARIMA model on the complex unit circle.

Usage

## S3 method for class 'Arima2'
plot(x, roots = TRUE, title = NULL, tick.lab = NULL, ...)

Arguments

x	An Arima2 object. This parameter is an object created using the function arima2().
roots	Would you instead prefer to plot the roots on a unit circle? Insert logical type here.
title	Title of plot
tick.lab	Time vector of numeric or character/string type.
...	Other parameters

Details

The output of this function is a ggplot object, so layers may be added to the output of this function using ggplot2's plus operator.

Value

⁠Arima 2⁠ plot, which is a ggplot2 object. Type of plot is indicated through roots parameter.

Examples

plot(arima(lh, order = c(1,0,1)))
plot(x = arima(lh, order = c(3,0,1)), roots = FALSE)

---

Profile for Arima2 object

Description

This function performs profile log-likelihood of an Arima2 function.

Usage

## S3 method for class 'Arima2'
profile(
  fitted,
  d = 0,
  npts = 100L,
  lower = -1,
  upper = 1,
  which = 1L,
  max_iters = 1,
  ...
)

Arguments

fitted	An Arima2 object that has been fit to data.
d	Integer number of differences. Should match the differences used to obtain the fitted object.
npts	Integer number of points to evaluate the profile.
lower	Numeric lower bound for the profile search.
upper	Numeric upper bound for the profile search.
which	Integer indicating which parameter to perform the profile over. See Details section for more information.
max_iters	Maximum number of random restarts. See arima for more details.
...	additional arguments needed for the profile function

Details

The parameter which specifies parameter in the following vector will be profiled over:

$\phi_1, \ldots, \phi_p, \theta_1, \ldots, \theta_q, \Phi_1, \ldots, \Phi_P, \Theta_1, \ldots, \Theta_Q, \mu$

where $p, q$ are non-negative integers representing the number of AR and MA coefficients of fitted, respectively, and $\phi_i$ are the AR coefficients, $\theta_i$ are the MA coefficients, $\Phi_i$ are the seasonal AR coefficients, $\Theta_i$ are the seasonal MA coefficients and $\mu$ is the model intercept.

Value

data.frame object containing the results of the profile likelihood.

Examples

# example code
set.seed(123)
mod <- arima(miHuron_level$Average, order = c(1, 0, 1), max_iters = 100)
prof <- profile(mod, which = 2L, lower = -0.5, upper = 0.5)
plot(prof$ma1, prof$loglik)

---

Sample ARMA coef

Description

This function randomly samples the ARMA coefficients of a specified ARMA model. The coefficients are sampled so that the resulting ARMA model is both causal and invertible.

Usage

sample_ARMA_coef(
  order = c(0L, 0L),
  seasonal = list(order = c(0L, 0L), period = NA),
  n = 1,
  Mod_bounds = c(0.05, 0.95),
  min_inv_root_dist = 0
)

Arguments

order	A specification of the non-seasonal part of the ARIMA model: this is different than the order input of stats::arima(), because the degree of differencing is not included. Thus order is a vector of length two of the form $(p, q)$.
seasonal	A specification of the seasonal part of the ARIMA model. Can be either a vector of length 2, or a list with an order component; if a list, a period can be included, but it does not affect the function output.
n	An integer indicating how many sets of ARMA coefficients should be sampled.
Mod_bounds	Bounds on the magnitude of the roots.
min_inv_root_dist	This parameter is included so as to help avoid ARMA models that contain parameter redundancy, if desired. Specifically, this parameter ensures that the minimum distance between any of the inverted roots in the AR and MA polynomials is greater than min_inv_root_dist. Inverted roots that are near each other leads to canceling or nearly canceling roots, effectively reducing the size of the ARMA model.

Details

For an ARMA model to be causal and invertible, the roots of the AR and MA polynomials must lie outside the the complex unit circle. The AR and MA polynomials are defined as:

$\phi(z) = 1 - \phi_1 z - \phi_2 z^2 - \ldots - \phi_p z^p$

$\theta(z) = 1 + \theta_1 z + \theta_2 z^2 + \ldots + \theta_q z^q$

where $z$ is a complex number, $\phi_1, \ldots, \phi_p$ are the $p$ AR coefficients of the ARMA model, and $\theta_1, \ldots, \theta_p$ are the $q$ MA coefficients of the ARMA model.

ARMA coefficients are sampled by sampling inverse roots to be inside the complex unit circle, and then calculating the resulting polynomial. To ensure that the resulting polynomial coefficients are real, we only sample half of the needed number of complex roots, and set the remaining half to be the complex conjugate of the sampled points. In the case where the number of coefficients is odd, the remaining root is sampled uniformly, satisfying the Mod_bounds parameter.

Value

a vector of randomly sampled ARMA coefficients.

Examples

{
sample_ARMA_coef(
   order = c(2, 1),
   seasonal = list(order = c(1, 0), period = 2),
   n = 100
)
}

---