Navigation

Operators and Keywords

Function List:

C++ API

tsa

Stochastic concepts and maximum entropy methods for time series analysis

Select category:

Univariate (stationary) analysis

acovf
ACOVF estimates autocovariance function (not normalized) NaN's are interpreted as missing values.
acorf
Calculates autocorrelations for multiple data series.
biacovf
BiAutoCovariance function [BiACF] = biacovf(Z,N);
bispec
Calculates Bispectrum [BISPEC] = bispec(Z,N);
durlev
function [AR,RC,PE] = durlev(ACF); function [MX,PE] = durlev(ACF); estimates AR(p) model parameter by solving the Yule-Walker with the Durbin-Levinson recursion for multiple channels INPUT:
lattice
Estimates AR(p) model parameter with lattice algorithm (Burg 1968) for multiple channels.
rmle
RMLE estimates AR Parameters using the Recursive Maximum Likelihood Estimator according to [1] Use: [a,VAR]=rmle(x,p) Input: x is a column vector of data p is the model order Output: a is
pacf
Partial Autocorrelation function [parcor,sig,cil,ciu] = pacf(Z,N);
parcor
estimates partial autocorrelation coefficients Multiple channels can be used (one per row).
invest0
First Investigation of a signal (time series) - automated part [AutoCov,AutoCorr,ARPMX,E,ACFsd,NC]=invest0(Y,Pmax);
invest1
First Investigation of a signal (time series) - interactive [AutoCov,AutoCorr,ARPMX,E,CRITERIA,MOPS]=invest1(Y,Pmax,show);
selmo
Model order selection of an autoregrssive model [FPE,AIC,BIC,SBC,MDL,CAT,PHI,optFPE,optAIC,optBIC,optSBC,optMDL,optCAT,optPHI]=selmo(E,N);
selmo2
SELMO2 - model order selection for univariate and multivariate autoregressive models X = selmo(y,Pmax); y data series Pmax maximum model order X.A, X.B, X.C parameters of AR mode
histo
HISTO calculates histogram for each column [H,X] = HISTO(Y,Mode) Mode 'rows' : frequency of each row '1x' : single bin-values 'nx' : separate bin-values for each column X are th
histo2
HISTO2 calculates histogram for multiple columns with separate bin values for each data column.
histo3
HISTO3 calculates histogram for multiple columns with common bin values among all data columns, and can be useful for data compression.
hup
HUP(C) tests if the polynomial C is a Hurwitz-Polynomial.
ucp
UCP(C) tests if the polynomial C is a Unit-Circle-Polynomial.
y2res
Y2RES evaluates basic statistics of a data series R = y2res(y) several statistics are estimated from each column of y OUTPUT: R.N number of samples, NaNs are not counted R.SUM sum
ar_spa
AR_SPA decomposes an AR-spectrum into its compontents [w,A,B,R,P,F,ip] = ar_spa(AR,fs,E);
detrend
DETREND removes the trend from data, NaN's are considered as missing values DETREND is fully compatible to previous Matlab and Octave DETREND with the following features added: - handles NaN's by
flix
floating point index - interpolates data in case of non-integer indices

Multivariate stationary analysis

mvar
MVAR estimates parameters of the Multi-Variate AutoRegressive model
mvfilter
Multi-variate filter function
mvfreqz
MVFREQZ multivariate frequency response [S,h,PDC,COH,DTF,DC,pCOH,dDTF,ffDTF,pCOH2,PDCF,coh,GGC,Af,GPDC,GGC2,DCOH] = mvfreqz(B,A,C,f,Fs) [...] = mvfreqz(B,A,C,N,Fs) INPUT: ======= A, B mult
arfit2
ARFIT2 estimates multivariate autoregressive parameters of the MVAR process Y
histo4
HISTO4 calculates histogram of multidimensional data samples and supports data compression

Adaptive (time-varying) analysis

aar
Calculates adaptive autoregressive (AAR) and adaptive autoregressive moving average estimates (AARMA) of real-valued data series using Kalman filter algorithm.
aarmam
Estimating Adaptive AutoRegressive-Moving-Average-and-mean model (includes mean term)
adim
ADIM adaptive information matrix.
amarma
Adaptive Mean-AutoRegressive-Moving-Average model estimation [z,e,ESU,REV,V,Z,SPUR] = amarma(y, mode, MOP, UC, z0, Z0, V0, W);
mvaar
Multivariate (Vector) adaptive AR estimation base on a multidimensional Kalman filer algorithm.

Conversions between forms

ac2poly
converts the autocorrelation sequence into an AR polynomial [A,Efinal] = ac2poly(r)
ac2rc
converts the autocorrelation function into reflection coefficients [RC,r0] = ac2rc(r)
ar2rc
converts autoregressive parameters into reflection coefficients with the Durbin-Levinson recursion for multiple channels function [AR,RC,PE] = ar2rc(AR); function [MX,PE] = ar2rc(AR);
rc2ar
converts reflection coefficients into autoregressive parameters uses the Durbin-Levinson recursion for multiple channels function [AR,RC,PE,ACF] = rc2ar(RC); function [MX,PE] = rc2ar(RC);
poly2ac
converts an AR polynomial into an autocorrelation sequence [R] = poly2ac(a [,efinal] );
poly2ar
Converts AR polymials into autoregressive parameters.
poly2rc
converts AR-polynomial into reflection coefficients [RC,R0] = poly2rc(A [,Efinal])
rc2ac
converts reflection coefficients to autocorrelation sequence [R] = rc2ac(RC,R0);
rc2poly
converts reflection coefficients into an AR-polynomial [a,efinal] = rc2poly(K)
ar2poly
converts autoregressive parameters into AR polymials Multiple polynomials can be converted.

Utility functions

arcext
ARCEXT extracts AR and RC of order P from Matrix MX function [AR,RC] = arcext(MX,P);
sinvest1
SINVEST1 shows the parameters of a time series calculated by INVEST1 only called by INVEST1
sbispec
SBISPEC show BISPECTRUM
flag_implicit_samplerate
The use of FLAG_IMPLICIT_SAMPLERATE is in experimental state.

Test suites

tsademo
TSADEMO demonstrates INVEST1 on EEG data
bisdemo
BISDEMO (script) Shows BISPECTRUM of eeg8s.mat
invfdemo
invfdemo demonstrates Inverse Filtering

Package: tsa