jbnorm_var1 Econometric tests and diagnostics ljungbox

Grocer >> Econometric tests and diagnostics > li_mcleod

li_mcleod

Li McLeod autocorrelation test

CALLING SEQUENCE

res = li_mcleod(namey,p,arg1,…,argn)

PARAMETERS

Input

• namey = a time series, a real (nx1) vector or a string equal to the name of a time series or a (nx1) real vector between quotes

• p = a scalar, the # of autocorrelations considered

• argi = arguments which can be:

 - 'noplt' if the user does not want to plot the autocorrelations

 - 'noprint' if the user does not want to print the results of the test

 - 'meth=Box-Pierce' if the user wants to use the Box-Pierce formula instead of the Li Mcleod one

 - 'size=x' if the user wants to choose the size of the autocorrelation confidence band (defaut = 0.05)

 - 'styleg=x' with x between 1 and 7 if the user wants to choose the location of the legend for the graph of autocorrelations (default = 7, that is to the right of the graph)

 - 'dropna' if the user wants to remove the NA values from the data 

Output

• res = a results tlist with:

  - res('meth') = 'Li Mcleod'

  - res('y') = values of the input variable (residuals in the case of a tlist)

  - res('acf') = autocorrelation coefficients

  - res('acf_l') = low bound of the confidence interval

  - res('acf_u') = upper bound of the confidence interval

  - res('size') = size of the confidence band

  - res('chistat') = Li Mcleod (or Box-Pierce) chi-squared

  - res('chi_pvalue') = the p-value of the test

  - res('chi_df') = the # of degrees of freedom

  - res('namey') = name of the variable tested

  - res('prests') = a boolean indicating whether the tested variable is a ts

  - res('bounds') = the estimation period (if the variable was a ts)

  - res('dropna') = boolean indicating if NAs have been droped

  - res('nonna') = vector indicating position of non-NAs (if the option 'dropna' was active)

DESCRIPTION

Computes the Li McLeod Q test of autocorrelation for a squared vector or the squared residuals of a results tlist.

EXAMPLE

load(GROCERDIR+'/macros/grocer/db/bdhenderic.dat'); 
bounds()
res1=li_mcleod('delts(lm1)',6)
//Example taken from function li_mcleod_d. Provides the Li Mcleod test for the variation of the logrithm of M1 in the UK and 1 to 6 autocorrelations.
r=varma_d4();
res2=li_mcleod(r,10)
// Also taken from function li_mcleod_d. Provides the Li Mcleod test for the residuals of the arma estimation provided in demo varma_d4() and 1 to 10 autocorrelations.

AUTHOR

Eric Dubois 2007
jbnorm_var1 Econometric tests and diagnostics ljungbox