TimeSeries
Utilities for multivariate time series
These utilities assume that the time series is stored in a Breeze DMD with variables/component series in columns, and observations in rows with the first row corresponding to the first observation and the last row to the last.
Attributes
- Graph
-
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
-
TimeSeries.type
Members list
Value members
Concrete methods
Compute the auto-covariance of the time series at a given lag
Compute the auto-covariance of the time series at a given lag
Value parameters
- centre
-
Does the time series need to be centred?
- lag
-
The lag at which the auto-covariance is to be computed (can be negative)
- x
-
Time series
Attributes
- Returns
-
Auto-covariance matrix
Compute a set of auto-covariance matrices
Compute a set of auto-covariance matrices
Value parameters
- centre
-
Does the time series need to be centred?
- lagMax
-
The maximum lag required
- x
-
Time series
Attributes
- Returns
-
List of
lagMax+1auto-covariance matrices (from lag 0 tolagMax)
The correlation matrix of a time series
The correlation matrix of a time series
Value parameters
- centre
-
Does the time series need to be centred?
- x
-
A time series
Attributes
- Returns
-
Correlation matrix
Compute the covariance between two time series of the same length
Compute the covariance between two time series of the same length
Value parameters
- centre
-
Do the series need to be centred, first?
- x
-
A time series
- y
-
Another time series
Attributes
- Returns
-
Covariance matrix
Fit a mean-zero VAR(p) model to a time series via least squares. Since the model being fit is mean zero, you may want to mean-centre your data.
Fit a mean-zero VAR(p) model to a time series via least squares. Since the model being fit is mean zero, you may want to mean-centre your data.
Value parameters
- centre
-
Mean-centre the time series before fitting?
- p
-
The order of the VAR(p) model to be fits
- x
-
Time series
Attributes
- Returns
-
The list of
pfitted VAR matrices, together with the lower Cholesky triangle of the estimated innovation variance matrix
Check whether a given set of VAR matrices (such as returned by fitVar) correspond to a VAR(p) model that is stationary
Check whether a given set of VAR matrices (such as returned by fitVar) correspond to a VAR(p) model that is stationary
Value parameters
- phi
-
List of VAR matrices
Attributes
- Returns
-
Is this model stationary?
Sweep out the mean of each component series
Sweep out the mean of each component series
Value parameters
- x
-
A multivariate time series
Attributes
- Returns
-
A mean zero time series
The variance matrix of a time series
The variance matrix of a time series
Value parameters
- centre
-
Does the time series need to be centred?
- x
-
A time series
Attributes
- Returns
-
Variance matrix