3
$\begingroup$

After testing for cointegration with Johansen test using the following code

[h,pValue,stat,cValue,mles] = jcitest(ret,'model','H1*','lags',4,'display','params')

I received this output:

Data: prices
Effective sample size: 56
Model: H1*
Lags: 4
Statistic: trace
Significance level: 0.05


r  h  stat      cValue   pValue   eigVal   
----------------------------------------
0  1  40.5214   35.1929  0.0161   0.3350  
1  0  14.6781   20.2619  0.1453   0.1667  
2  0  5.4670    9.1644   0.1580   0.1091 

So it rejects no cointegration. So I am running the VECM with rank 1 restriction:

B = mles.r1.paramVals.B % Cointegrating relations with rank = 1 restriction

I get the following output:

r = 1
------
A =
-0.6259
-0.2261
-0.0222

B =
0.7081
1.6282
-2.4581

B1 =
0.0579 1.0824 -0.8718
0.1182 0.4993 -0.5415
0.1050 0.1667 -0.1600

B2 =
-0.5462 2.2436 -1.7723
-0.1518 0.6605 -0.6169
-0.1610 0.5966 -0.5366

c0 =
2.2351

c1 =-0.0366
0.0872
0.1444

How would I interpret this output and how do I move forward after estimating the parameters? (I have only displayed the r=1, and these are not the actual results, by I couldn't find any explanation behind them)

$\endgroup$
2
$\begingroup$

Based on the official documentation, you have found that there is one cointegration relationship with 95% probability. Given your selection of Model: H1* and Lags: 4, the model you are estimating is:

$$ \Delta y_t = A(B´y_{t−1}+c_0) + B_1\Delta y_{t-1} + B_2\Delta y_{t-2} + B_3\Delta y_{t-3} + B_4\Delta y_{t-4} + \epsilon_t $$

The matrix B is of dimension $n \times r$, and it contains as columns the cointegration vector(s). In your case, $r=1$, and the cointegration vector is

B =
0.7081
1.6282
-2.4581 

The Matrix A represent the "error-correction" matrix. For a system to converge towards the equilibrium cointegration relation (i.e. to be stable), you want these numbers to be in between -2 and 0 (as in your case).

c0 is just a constant of the cointegration relationship (like in a standard function $y=c0+bx$).

The matrices Bi show how previous period shocks translate through the system over time.

$\endgroup$
  • $\begingroup$ Thank you. What about the other paramters? Reading other posts a value greater than +- 2 is significant. Is this the right assumption? $\endgroup$ – Adrian Aug 26 '17 at 10:10
  • $\begingroup$ @Adrian Can you a bit more specific on which parameters are you refering to? Not sure what you mean. $\endgroup$ – luchonacho Aug 26 '17 at 17:47
  • $\begingroup$ A, B, B1, B2, c0, c1 $\endgroup$ – Adrian Aug 27 '17 at 9:21
  • $\begingroup$ @Adrian what do you think of the update? Still got questions? $\endgroup$ – luchonacho Sep 1 '17 at 10:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.