QUICK START GUIDE
Time Series Analysis with MATLAB® and Econometrics Toolbox™
This reference shows common use cases but is not comprehensive.
The icon provides a link to relevant sections of the documentation.
Analyze Time Series Using the Econometric Modeler App
Import data
Conduct a specification test
Transform data
Select models
Perform model diagnostics
Share session results:
• Export variables
• Generate functions and reports
Econometric Modeler App Overview
Data Transformation Data Visualization
Prices Returns ACF Plot PACF Plot Correlation Plot
Returns = price2ret(Prices); autocorr(y); parcorr(y); corrplot(X);
Prices = ret2price(Returns);
First-Order Differencing
dy = diff(y);
∆𝑦𝑡= 𝑦𝑡− 𝑦𝑡−1
Detrending
y = detrend(y);
Model Comparisons
Akaike or Bayesian Information Criteria
Learn more: mathworks.com/help/econ
[aic,bic] = aicbic(logL,numParam,numObs);
mathworks.com
Specification Tests Conditional Mean Models
ARMA, ARIMA, and ARIMAX
Stationarity
Create Models
[h,pValue] = testName(y);
Mdl = arima(p,D,q);
adftest Augmented Dickey-Fuller test
kpsstest KPSS test for stationarity Estimate/Fit
lmctest [EstMdl,EstParamCov,logL,info] = estimate(Mdl,Y);
Leybourne-McCabe stationarity test
pptest Phillips-Perron test for one unit root Impulse Simulate Forecast
vratiotest Variance ratio test for random walk impulse(Mdl); [Y,E] = ... [Y,YMSE] = ...
simulate ... forecast ...
(EstMdl,numObs); EstMdl,numperiods,Y0);
Heteroscedasticity
Engle test
[h,pValue] = archtest (residual);
Correlation
Ljung-Box Q-Test for autocorrelation
[h,pValue] = lbqtest(residual);
Belsley Collinearity Diagnostics
collintest(X) Multivariate Model
Vector Autoregression (VAR)
Cointegration and Vector-Error Correction (VEC)
egcitest Engle-Granger cointegration test Create Models
jcitest VAR
Johansen cointegration test
Mdl = varm(numseries,numlags);
jcontest Johansen constraint test
VEC
Causality Mdl = vecm(numseries,rank,numlags);
Block-wise Granger causality and block exogeneity tests
[h,pValue] = gctest(Y1,Y2); Estimate/Fit
[EstMdl,EstSE,logL,E] = estimate(Mdl,Y);
Conditional Variance Models Investigate
GARCH, EGARCH, and GJR h = gctest(Mdl); % For VAR model only
Create Models Response = irf(Mdl);
Mdl = garch(p,q); Decomposition = fevd(Mdl);
Mdl = egarch(p,q); Simulate
Mdl = gjr(p,q); [Y,E] = simulate(EstMdl,numObs);
Estimate / Fit
[EstMdl,EstParamCov,logL,info] = estimate(Mdl,Y);
Simulate Forecast
[V,Y] = V =
simulate(EstMdl,numObs); forecast(EstMdl,numperiods,Y0);
Forecast
[Y,YMSE] = forecast(EstMdl,numperiods,Y0);
mathworks.com
© 2019 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See mathworks.com/trademarks for a list of additional trademarks.
Other product or brand names may be trademarks or registered trademarks of their respective holders.
