10.11.2014 Views

CES VŠEM o aktuálních ekonomických problémech

CES VŠEM o aktuálních ekonomických problémech

CES VŠEM o aktuálních ekonomických problémech

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

1. all coefficients are tested via t-test;<br />

2. test, where or;<br />

3. the joint hypothesis that under the<br />

and the last components of are zero. 36<br />

As Cuestas (2009) mentions, rejection of the left side hypothesis<br />

for the first and second tests means linear or non-linear trend<br />

stationarity. (It cannot be decided which of the cases is true.)<br />

Rejection of the right side means non-linear trend stationarity.<br />

The third test is a one-sided test that does not give us any answer<br />

regarding the trend as right side rejection is only the rejection of<br />

the . All possibilities are summarised in the table 1.<br />

Table 1: Alternative hypotheses (<br />

) for the Bierens (1997) test<br />

Test Left-side rejection<br />

Right-side rejection<br />

ST, TST or NTST NTST<br />

ST, TST or NTST NTST<br />

- ST, TST or NTST<br />

Note: ST – (mean) stationary, TST – (linear) trend stationary, NTST –<br />

non-linear (trend) stationarity. The test is only a one-sided test,<br />

the and are specified for both sides. Source: Cuestas, 2009,<br />

p. 92, own adaptation.<br />

1.8 Panel unit root tests<br />

The panel unit root tests are extensions of the URTs. Due to their<br />

construction, they offer more than a simple inclusion of crosssectional<br />

dimension. They differ as to what transformations are<br />

used and they assume linearity of the underlying process. All<br />

commonly used panel unit root tests assume the presence of<br />

36 An alternative is a test proposed in [Bierens (1997)] with a standard null<br />

distribution. The text also introduces three tests that are independent of the<br />

assumed AR structure for the error term .<br />

81

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!