Quantcast
Channel: EViews
Viewing all 24077 articles
Browse latest View live
↧

Sum of Squares

August 29, 2016, 9:20 am
↧
Search

Hausman test: different results for different components estimator

August 29, 2016, 9:40 am
$
0
0
I am testing a number of random effects equations for possible correlation between the random effects and the independant varables. I am using the Hausman test in Eviews. My problem is that I get widely different results depending upon which estimator for components variance I have used. Swamy-Arora signals the existence of correlation for almost all the equations. Wansbeek-Kapteyn does not confirm correlation for any of the equations. And Wallace-Hussein reports that the test result is invalid for quite a few of the equations and gives a clear signal for the other equations: no statistically confirmed correlation.
So my question is: is there any arguments for putting more trust in any of these results than in the other two? Or should I choose the FE model just to avoid any risk?

↧
↧

Time varying SVAR

August 29, 2016, 9:42 am
$
0
0
Hi, Thank Dakila for help on my first question, I have downloaded the add-in, that looks ok, however when following document instructions, I put the details in it gives message "Near singualr Matrix". I don,t get any impulse response but some series and matrixes appear when do not make much sense to me.Please help.

Regards

↧

Time varying SVAR

August 29, 2016, 10:28 am
$
0
0
the file contains three variables g taxes and gdp all in log and first difference
tvp var .wf1


↧

Sum of Squares

August 29, 2016, 12:42 pm
↧
↧

Sum of Squares

August 29, 2016, 1:44 pm
↧

Sum of Squares

August 29, 2016, 1:50 pm
$
0
0
I can't figure out the programming, could you help me out?

↧

IRF with VARMA GARCH

August 29, 2016, 5:02 pm
$
0
0
Dear All,

My first attempt at modeling VARMA GARCH models with Eviews is turning ugly. In particular, I need to get Impulse Response Functions with bootstrapped confidence intervals.

As I do not see anything in Eviews for this, I am thinking about a two-stage process. First, to bootstrap/simulate the VARMA process, then to do the same with residuals, and then to add up the lower and upper bounds to get the right simulated confidence intervals.

That doesn't look stupid as everything is linear, but somehow that does feel ugly and dubious.

Any suggestion to get around this issue?

Thanks in advance,

Superleo

↧
Search

Time varying SVAR

August 30, 2016, 7:44 am
$
0
0
There is a bug. I will try to fix it. However you can use the following code to fix it:
Code:
pagestruct(start=1998q2)


↧
↧

Time varying SVAR

August 30, 2016, 7:48 am
$
0
0
ali_Economist wrote:Hi, Thank Dakila for help on my first question, I have downloaded the add-in, that looks ok, however when following document instructions, I put the details in it gives message "Near singualr Matrix". I don,t get any impulse response but some series and matrixes appear when do not make much sense to me.Please help.

Regards


1. Try to transform your variables. For example, to annual growth rate
2. Try to change the default prior parameters.

↧

Hansen instability test after DOLS estimation

August 30, 2016, 8:52 am
$
0
0
Dear all,

Probably I'm wrong, but I have the feeling that there is a problem with the Hansen instability test after a DOLS estimation of cointegration parameters.

Unlike with the test performed after a FM-OLS regression, it seems to me that after a DOLS regression the test cannot never reject the null of cointegration (stability). I obtain the same result even if I completely misspecify the cointegrating regression. Indeed, the test statistic is always extremely low, such that the null is never rejected. Please note that other cointegration tests do reject the null of no cointegration.

Below you will find the ouput from the same regression using DOLS or FM-OLS. The estimated parameters are a bit different, but I can hardly believe that this difference explains the divergence in the output of the Hansen test.

Thank you very much for your help.

Mat


Dependent Variable: LOG(M1RSY)                
Method: Dynamic Least Squares (DOLS)
Included observations: 105 after adjustments                
Cointegrating equation deterministics: C                
Fixed leads and lags specification (lead=4, lag=4)                
HAC standard errors & covariance (Bartlett kernel, Newey-West automatic                
bandwidth = 9.0000, NW automatic lag length = 4)                
                
Variable    Coefficient    Std. Error    t-Statistic    Prob.
                
LOG(R)    -0.350357    0.059621    -5.876414    0.0000
C    -2.612783    0.245019    -10.66358    0.0000
                

Cointegration Test - Hansen Parameter Instability                
Series: LOG(M1RSY) LOG(R)                
Null hypothesis: Series are cointegrated                
Cointegrating equation deterministics: C                
HAC score variance                
                
    Stochastic    Deterministic    Excluded    
Lc statistic    Trends (m)    Trends (k)    Trends (p2)    Prob.*
0.004220     1     0     0    > 0.2
                
        
                
Dependent Variable: LOG(M1RSY)                
Method: Fully Modified Least Squares (FMOLS)                
Included observations: 113 after adjustments                
Cointegrating equation deterministics: C                
Long-run covariance estimate (Bartlett kernel, Integer Newey-West                
automatic bandwidth = 9.0000, NW automatic lag length = 4)                
                
Variable    Coefficient    Std. Error    t-Statistic    Prob.
                
LOG(R)    -0.219729    0.083819    -2.621476    0.0100
C    -2.177911    0.300903    -7.237923    0.0000
                
R-squared    0.114549     Mean dependent var        -1.454140
Adjusted R-squared    0.106572     S.D. dependent var        0.363920
S.E. of regression    0.343982     Sum squared resid        13.13391
Long-run variance    0.729154            
                

Cointegration Test - Hansen Parameter Instability                
Series: LOG(M1RSY) LOG(R)                
Null hypothesis: Series are cointegrated                
Cointegrating equation deterministics: C                
                
    Stochastic    Deterministic    Excluded    
Lc statistic    Trends (m)    Trends (k)    Trends (p2)    Prob.*
1.192470     1     0     0    < 0.01

↧

Time varying SVAR

August 30, 2016, 9:56 am
$
0
0
Ok I will try by resizing the simple. thank u very much dakila

↧

Misleading calculations of Endogeneous variables in system

August 30, 2016, 10:19 am
$
0
0
Dear Gareth,

Thanks a lot again. I now agree - not only by your little example but by a slightly more complex one I set up and by looking into old work files of mine which showed the same behaviour.

However, and I know how strange that may sound and that I may look quite silly - I am the more sure that my current work file does not do this the way it is supposed to (actually it does it the other way round, meaning if the edogeneous explaining variable is not excluded the model calulates the dependent equation on the actuals and not the estimates, and only if it is excluded after a first run of the model, the model then takes the estimated [and aliased] variable in a second run and calculates the estimates on the estimated values).

So, I quess my file is somehow corrupted, no other explanation.

Thanks for your support.
Christian

↧
↧

first value

August 30, 2016, 11:47 am
$
0
0
When setting sample is there a way to have the sample start when the series starts?

For instance, instead of writing "smpl 1989Q2 2016Q3" after having checked the series start date, I would like to write a code that checks the series start date itself.

Please let me know if I'm not clear.

Thanks for your help.

Trainee

↧

first value

August 30, 2016, 11:53 am
$
0
0
In one line:
Code:

smpl @first 2016q3 if @date >= @dateval(X.@first)


Personally I would probably do something like this though:
Code:

%first = x.@first
smpl {%first} 2016q3


which is a bit easier to read.

Note neither will be dynamic - the sample will not automatically adjust if the series start date changes.

↧
Search

Gregory-Hansen and Threshold Cointegration

August 28, 2016, 12:50 am
$
0
0
Hello,

I have used the Johansen multivariate cointegration test to see whether a group of ten stock markets are cointegrated. However, I was wondering whether it would be possible to use the Gregory-Hansen and Enders-Siklos tests for structural-breaks and threshold adjustment on the entire system or whether they can only be used in a pairwise manner?

Any help is appreciated.

↧

Gregory-Hansen and Threshold Cointegration

August 30, 2016, 6:19 pm
$
0
0
Yes, it is possible. Just try it

↧
↧

Pausing a program

August 31, 2016, 6:19 am
$
0
0
This is a somewhat different question than the previous ones, but I put it here anyway.

If I run a Monte Carlo simulation program that takes me e.g. 10 hours, and I would like to temporarily pause the simulation anywhere in the middle (since maybe I need the computer for something else that requires processor power), and then later I want to resume the same simulation for remaining hours. Is that possible? (I do not want to push escape and start from the beginning if that is possible, I want to resume after the break). /P

↧

syntax cloroing

August 31, 2016, 6:51 am
$
0
0
Hi

after the update, comments in program files do not have a green color..? see picture

syntax coloring.JPG


why is this?
Thomas

↧

syntax cloroing

August 31, 2016, 6:54 am
$
0
0
when i write a new comment, it becomes green. but when i open a program file with comments, the exising comments in that file are not green

↧
Viewing all 24077 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>