I need to take the log difference of a matrix, i.e. log(M[2 rows(M):,]/M[1:rows(M)-1,]). Unfortunately, M has elements equal to zero. I need to replace the nan's and inf's with 0's. This almost works
M = isnan(M) ? 0 : M
but does not remove inf's. Any sugestions?
I am writing to seek help to forecast volatility of index returns using
In GRETL, I want to know how to perform in the sample and out of sample
forecasting after estimating index return series by using GARCH variants
in the gig package.
I'm currently estimating a bootstraped bivariate probit through a
progressive loop and retrieve each time the $yhat matrix. At some point,
the execution is interrupted with the "warning":
The statistic you requested is not available
>> genr series predict_external = $yhat[,1]
Is there a perfect prediction symptom behind this message isn't it?
Can one, in such a loop, skip cases where the MLE estimation is not
Dear Mr. Schreiber,
since 1 year and a half I'm writing a book with the german title "Einführung in die Ökonometrie mit Gretl". It's an introduchtion into the theory of econometrics and I tried to connect the conceptions with a direct conversion into Gretl-Skripts/Dialogs etc. This approach has the advantage, that the student does not only learn the theory but also the basic concepts of Gretl. In the most books which concern with econometrics the content is limited on the theory and perhaps on Output-listings.
I will finish the work in the next days and hope to find a publisher. Please notice the content of my work in the attachement... It's written in german and unfortunately not in english. Perhaps it will nevertheless find your pleasure.
The work is about 540 pages and many themes are considered, as you will see....
I take refer to the Gretl-conference next year in Berlin... Perhaps it's an occasion to present the book there ??
Please forgive my modest skills in the English language... If you wish I translate the content into English.
What are the 3 "model selection statistics" that Gretl mentions when doing tests omitting variables? I'm guessing adj. R-squared, log-likelihood & AIC, but what about SC & HQ? Or is she referring to the 3 info criteria?
Please keep the communication on the list. [See below for Karthik's
Am 29.09.2014 um 15:46 schrieb Karthik Raju:
> Hello Sven,
> First, I wish to congratulate you for organizing the 4th GRETL
> conference to be held at Berlin on 12 & 13 June 2015.
> Now, I will cite you a literature related to what I intend to do.
> Please refer to the article in the link http://goo.gl/zwhzS7. In this
> article, the author compare the predicted variance from different models
> to the realized variance and evaluates the performance of those models
> by using loss functions to choose a best model.
> Similarly, I have a dataset comprising 5 years of stock index returns
> for which I want to estimate the conditional variance using models like
> TARCH, EGARCH, APARCH etc. Then, I want to forecast/predict the models
> for 1 year ahead and measure the forecasting/predicting ability of the
> models used.
> Karthik Raju
> Sven Schreiber wrote on 2014-09-29 10:50 AM +0530:
>>> Please show us some example from the literature where this is done, in
>>> order to address the conceptual difficulties that Jack mentioned. Only
>>> afterwards could we possibly move to the stage of formulating code.
>> Am 28.09.2014 um 10:26 schrieb Karthik Raju:
>> I am a beginner in time series analysis. So, please can you guide me how
>> to formulate the code for unique models like EGARCH, TARCH etc.
>> Gretl-users mailing list
I am writing to seek help in interpreting results of GARCH variants and
to forecast volatility in index returns.
I am a beginner in time series analysis, currently working on analyzing
the time varying volatility in index returns. With reference to the
logarithmic index returns can anyone put light on,
1) How to interpret the output of a GARCH variants?
2) How to forecast variance with the GARCH outputs?
3) How to calculate loss functions to interpret forecasting performance
of GARCH variants used?