Search found 1518 matches

by trubador
Sat Nov 05, 2016 6:11 pm
Forum: Bug Reports
Topic: Null hypothesis in the omitted variables test
Replies: 0
Views: 1133

Null hypothesis in the omitted variables test

I believe the null hypothesis in the omitted variables test should either read "..are NOT jointly significant..." or "...are jointly INsignificant..." wfcreate u 100 series x1 = nrnd series x2 = nrnd series y = 2 + .8*x1 + 0.2*x2 + @rnorm equation eq1.ls y c freeze(omitted) eq1.t...
by trubador
Fri Nov 04, 2016 12:53 am
Forum: Add-in Support
Topic: L1FILTER
Replies: 0
Views: 2203

L1FILTER

This thread is about l1filter add-in. The add-in allows you to extract a piecewise linear trend by primal-dual interior point method proposed by Kim et al. (2009). It is an alternative trend estimation method to well-known Hodrick-Prescott filter, but can also be used in determining the break (kink)...
by trubador
Sun Jul 17, 2016 3:44 pm
Forum: Suggestions and Requests
Topic: EViews 10
Replies: 16
Views: 5264

Re: EViews 10

I am not sure if the seed option would be enough to produce random subsets for k-fold CV as each set should hold unique values of the total sample (e.g. draw without replacement). It should be clear that the code above does not ensure this property.
by trubador
Sun Jul 17, 2016 9:31 am
Forum: Suggestions and Requests
Topic: EViews 10
Replies: 16
Views: 5264

Re: EViews 10

I think OP refers to me as the "author" who put the effort in writing the k-fold CV subroutine: http://forums.eviews.com/viewtopic.php?f=23&t=12261 I already shared with him the code behind this routine and totally agree that such approach (i.e. cross validation) should be made availab...
by trubador
Wed Jul 13, 2016 11:40 am
Forum: Econometric Discussions
Topic: Root Mean Square Error in Forecasting
Replies: 1
Views: 1067

Re: Root Mean Square Error in Forecasting

That's the way how RMSE (or any other evaluation metric) is computed. There is nothing inherently wrong here. Whether or not to keep insignificant variables in the model is up to you (based on the research question, hypotheses and other statistical criteria of course).
by trubador
Wed Jul 13, 2016 11:32 am
Forum: Add-in Support
Topic: DCCGARCH11
Replies: 117
Views: 56208

Re: DCCGARCH11

Hey, So right now I am estimating bivariate dcc garch (return in 2 indices such as s&p500 and klse). Some of the paper that I read, define the mean equation such: r_t=γ_0 + γ_1 r_(t-1)+ γ_2 r_(t-1)^us + ε_t ( I attached the paper's method and result) , it uses ar(1) process in the mean equation...
by trubador
Wed Jul 13, 2016 8:17 am
Forum: Add-in Support
Topic: HDecomp (historical decomposition)
Replies: 30
Views: 27026

Re: HDecomp (historical decomposition)

Hi Trubador, When using the historical decomposition, I am looking to identify the SVAR using the Choleski Decomposition which you have provided However, there does not appear to be any place to provide the casual ordering necessary to identify the SVAR using the Choleski Decomposition approach. (T...
by trubador
Wed Jul 13, 2016 8:10 am
Forum: Add-in Support
Topic: RunsTest
Replies: 4
Views: 3955

Re: RunsTest

From the documentation that comes with the add-in: Analysis returns a vector of two values. First value corresponds to the number of runs and the second value represents the associated p-value. As for computing the z-stat: You can exploit the following relationship: scalar zstat = @qnorm(1-pval/2)
by trubador
Wed Jun 22, 2016 4:34 pm
Forum: Add-in Support
Topic: TARCOINT
Replies: 46
Views: 31332

Re: TARCOINT

TARCOINT add-in is now updated to version 1.1 as of June 23, 2016.

Fixed a minor bug causing momentum option not to work properly when using the command line syntax. Supporting files are provided to guide users through some post-estimation analyses.
by trubador
Wed Jun 22, 2016 4:17 pm
Forum: Estimation
Topic: Interpreting impulse response functions: Std dev or % ?
Replies: 14
Views: 9246

Re: Interpreting impulse response functions: Std dev or % ?

In the first case, the interpretation of the impact should be in percentages since the response variables are all in logarithms. One SD shock to UNCER leads to a 0.004 units decrease in the logarithm of LIP after 6 months, which corresponds to 0.4% drop in the LIP when translated into original level...
by trubador
Wed Jun 22, 2016 5:55 am
Forum: Estimation
Topic: Interpreting impulse response functions: Std dev or % ?
Replies: 14
Views: 9246

Re: Interpreting impulse response functions: Std dev or % ?

Be careful that responses are always in the original variables and they are in whatever are the units of the variable. You cannot interpret the responses as percentage changes unless the data are put into the model in logs. In your case, response variables are log differenced, but you are interested...
by trubador
Wed Jun 22, 2016 3:47 am
Forum: Add-in Support
Topic: Spectral Granger Causality Test*
Replies: 12
Views: 7190

Re: Spectral Granger Causality Test*

I am not familiar with the methodology, but as I understand it does not have any problem with nonstationarity unlike its traditional time series counterpart. If I am not mistaken, this is one of the main advantages of working in the frequency domain. In any case, thanks for the effort and sharing.

Go to advanced search