CATS is an add-on program to RATS: Regression Analysis of Time Series , the cointegration facilities in Microfit, and a beta version of PC-FIML 8 is. By David Tufte; CATS in RATS: cointegration analysis of time series: version . CATS in RATS: Cointegration Analysis of Time Series. Front Cover. Henrik Hansen, Katarina Juselius. Estima, – Cointegration – 87 pages.
|Published (Last):||28 August 2015|
|PDF File Size:||1.73 Mb|
|ePub File Size:||13.98 Mb|
|Price:||Free* [*Free Regsitration Required]|
The test is distributed as 2 with p1 r degrees of freedom. The vectors are normalized so that VO 0 S11 VO D I p1 which is convenient for the analysis of the model properties but is most likely to have no economic interpretation.
First, you specify the restrictions, typically by inputting a number of design matrices. Hence, H and H impose 1 2 two over-identifying restrictions on each -vector which is why the degrees of freedom is 4 and not 6 as one might think.
The graphical and residual analysis are described in the following sections, while the recursive estimation procedures for investigating constancy of the parameters are demonstrated in chapter 4. The Automated Tests menu.
Here, we choose to restrict CQ 2 such that only the rst permanent shock, u tP1 ; has a lasting effect on the US bond rate. If restrictions are imposed on the vectors ofthe model 1.
The dialog has 6 menu items: The mentioned technical condition on A rules out variables integrated of higher order. Reports can now be reformatted on screen changing the displayed precisionand then exported or copy-and-pasted cointegratiin a number of formats. Testing Identifying Restrictions on The relations given by 0 X t are at most integrated of order rars and cointegrate with the differences 1X t to stationarity. NA CT 3 The test for normality is the Doornik-Hansen test, see Doornik and Hansen Overlaid on the histogram is the estimated density function of the standardized residuals grey line and the density of the standard normal distribution.
Hence, the shift is not necessary for accepting stationarity of the relation 3. We choose None to test the exogeneity restriction without imposing any restrictions on: When the recursive estimation has been performed you can always access this dialog either by selecting I I 1: The response on a shock to the German bond rate is higher prices in Germany and a small, probably insigni cant, effect on the US prices.
This implies that H. Often, a stationary variable might a priori play an important ccointegration in a hypothetical cointegration relation; for instance, an in ation rate. Should your model, for example, also contain exogenous and dummy variables, CATS will use the second supplementary card to set the exogenous variables, and the third to de ne the dummies. See also the descriptions of R and S1. This hypothesis is described in paragraph 3. The handbook describes how to use CATS 2.
Note that this way of specifying restrictions on does not allow for non-zero restrictions between the variables: The test statistic is asymptotically 2 distributed with.
If you add more xed relations it is not strictly necessary to generate the set anew since non-identifying structures will be ruled out anyway. If your model has catz cointegrating relations, xing a part of them can signi cantly reduce the computa- tion time for the model selection procedure.
The Multivariate Cointegration Model we obtain a model that allows for linear trends in all directions, also in the multi-cointegrating relations de ned below, while it excludes quadratic and cubic trends.
Simulate Critical Values simulates the critical values of the asymptotic distribution of the rank test. Before anything is printed, you are asked to normalize CQ 2 column-wise ; and B and B 1 row-wise as shown in gures 3. By default, CATS suggests a minimal base sample chosen as a function of number of parameters in the model.
This option will update mycats. Dialog for specifying restriction on of the form R 0 D 0.
Katarina Juselius – Google Scholar Citations
Select which relations you wish to include in the procedure and close the window. All the procedures eeries work the same way: Should the restrictions not formally identify the cointegrating space, the dialog box shown in gure 3. The second permanent shock, identi ed as a PPP shock, has by de nition no long-run impact on the US bond rate. Dialog for switching between the forms of restricting: This corresponds to the result in section 3.
This menu is only available after CATSmining has been run. The third permanent shock is identi ed as a shock to the German bond rate, and by de nition it has no permanent impact on the US bond rate and the pppt.
Printed in the USA.
TStat – O F F I C I A L – S I T E
MA Representation will print the matrices of the moving average representa- tion 1. This applies only to restrictions of the form R 0 D 0; recursive estimation of models with restricted as D. Here you can specify a number of settings controlling the procedure.
The default lag length is k D 2.