Depuis que Diebold & Inoue (2001), Granger & Hyung (2004) ou encore Haldrup & Nielsen (2007), ont mis en exergue les risques de confusion entre mémoire longue et break structurel, de nombreux tests ont été développés pour discriminer les deux phénomènes. Il est ici programmé un test récent proposé par Perron & Qu (2010):
The following RATS code is a source file.
Procedure LMSB
***** *****
* *
* ******************** Gilles de Truchis ******************* *
* *********** D1 Rech DEFI | UnivMed (U2) ************ *
* ************ 2010/2011 ************ *
* *
**** *****
********
*Compute the Perron & Qu (2010) Long Memory vs Structural Break Test
********
***************************************** LMSB function definition
procedure lmsb resids sbeg send
type series resids
type integer sbeg send
option real a 0.5
option real b 0.8
option real d1
option real d2
local vect[real] dvec
dim dvec(2)
if(%defined(d1).and.%defined(d2))
{
comp dvec(1) = d1, dvec(2) = d2
}
else
{
@gphlp(noprint,power=a) resids sbeg send
comp dvec(1) = %%d
@gphlp(noprint,power=b) resids sbeg send
comp dvec(2) = %%d
}
comp %pqstat = (sqrt(24*(send**0.5))/(%pi**2))*(dvec(1)-dvec(2))
comp %pqsignif = %ztest(%pqstat)
end
Sources:
Perron, P., Qu, Z., 2010. Long-Memory and Level Shifts in the Volatility of Stock Market Return Indices. Journal of Business & Economics Statistics, 28(2), 275-290.
Diebold, F. X., Inoue, A., 2001. Long memory and regime switching. Journal of Econometrics, 105, 131-159.
Granger, C. W. J., Hyung, N., 2004. Occasional structural breaks and long memory with an application to the S&P 500 absolute stock returns. Journal of Empirical Finance, 11, 399-421.
Haldrup, N., Nielsen, M.., 2007. Estimation of fractional integration in the presence of data noise. Computational Statistics & Data Analysis, 51, 3100-3114.
