Vous êtes sur la page 1sur 9

MATLAB toolboxes

June 2010

Jozef Rudy
J.Rudy@2009.ljmu.ac.uk
Liverpool John Moores University
1
Contents
• Description of TOOLBOXES and how to use
them

• Some free TOOLBOX and functions


suggestions

2
Description of TOOLBOXES (I)

• TOOLBOX is just the group of functions in some


specific area. E.g. econometrics, neural networks,
data acquisitions etc.

• The MATLAB version which you get at University


does not contain all the toolboxes

• However, there are some free TOOLBOXes on the


internet, which are many times better than the
original ones

3
Description of TOOLBOXES (II)

• When you download the TOOLBOX, you need


to enable it in MATLAB

• Go to FILE -> ADD PATH -> ADD WITH


SUBFOLDERS

• MATLAB will be aware of TOOLBOX functions


from now on
4
Econometrics (I)
• http://www.kevinsheppard.com/wiki/MFE_To
olbox

• To learn how to use functions, there is a


manual:
http://www.kevinsheppard.com/images/9/95/
MFE_Toolbox_Documentation.pdf

5
Econometrics (II) - stationarity
• Download any share from yahoo

• Test of stationarity – maximum number of significant lags will be selected

maxlag = 20;
[ADFstat, ADFpval, critval,resid, lags] = augdfautolag(x,1,maxlag,’AIC’);

6
Econometrics (III) - GARCH
• Download any share from yahoo

• Test for autocorrelation – sacf(returns.^2,20) – 20 is number of lags

• Estimation of GARCH(1,1) model – parameters = tarch(returns, 1,0,1)


• Estimation of GJR-GARCH(1,1) model – parameters = tarch(returns, 1,1,1)

• Test of stationarity – maximum number of significant lags will be selected


maxlag = 24;
[ADFstat, ADFpval, critval,resid, lags] = augdfautolag(x,1,maxlag,’AIC’);

7
Some other free toolboxes
• http://www.spatial-econometrics.com/ -
econometrics

• http://code.google.com/p/bnt/ - bayes net


toolbox – contains:
• Static
– Linear regression, logistic regression, hierarchical mixtures of experts
– Naive Bayes classifiers, mixtures of Gaussians, sigmoid belief nets
– Factor analysis, probabilistic PCA, probabilistic ICA, mixtures of these models
• Dynamic
– HMMs, Factorial HMMs, coupled HMMs, input-output HMMs, DBNs
– Kalman filters, ARMAX models, switching Kalman filters, tree-structured Kalman filters, multiscale AR models

• Many others – depending on the purpose of


use
8
Resources used
• www.Mathworks.com

• http://www.spatial-econometrics.com/

• http://www.kevinsheppard.com/wiki/MFE_Toolb
ox

• http://code.google.com/p/bnt/
[all accessed 2nd June 2010]
9

Vous aimerez peut-être aussi