Search found 15 matches

by Thomasms90
Tue Aug 23, 2016 9:49 pm
Forum: Estimation
Topic: Parameter restrictions in Switching Regressions
Replies: 5
Views: 5062

Re: Parameter restrictions in Switching Regressions

I wrote a bunch of if's to compare the regimes and correctly classify them in a table, then used that table for lookups.
by Thomasms90
Mon Jul 27, 2015 10:58 am
Forum: Data Manipulation
Topic: Format zeroline
Replies: 7
Views: 10953

Re: Format zeroline

Okay. It should be okay with a regular line if it's possible to get it from end to end, I just haven't been able to. Can you explain how?
by Thomasms90
Sun Jul 26, 2015 6:35 am
Forum: Data Manipulation
Topic: how to limit sample to a qualitative characteristics
Replies: 3
Views: 3564

Re: how to limit sample to a qualitative characteristics

Assuming your series "country" is an alpha series: series country_num = @iif(@eqna("USA", country)=1, 1, 2) Would create new numeric series and assign 1 to "USA" and otherwise 2 based on an alphanumeric string. Note you can't assign values to an alphanumeric series (i.e...
by Thomasms90
Sun Jul 26, 2015 3:19 am
Forum: Data Manipulation
Topic: Format zeroline
Replies: 7
Views: 10953

Format zeroline

Is it possible to format the zeroline? Ideally I'd like it in the background in light grey, dashed and 0.25 width. I can't find any option to do this and tried specififying some parameters after the -dashline parameter in my program, but it doesn't help. I also tried creating an extra series that's ...
by Thomasms90
Fri Jul 17, 2015 7:04 am
Forum: Bug Reports
Topic: Switchreg smoothed probabilities
Replies: 1
Views: 2473

Switchreg smoothed probabilities

Hi, Version: Eviews 8.1, standard edition, Apr 27 2015 build, 64-bit on Win7 Professional I extract smoothed probabilities from a Switching regression using the command: eq.makergmprobs(type=smooth) series1 series2 series3 When I open one of the series, in the top pane of the spreadsheet it says tha...
by Thomasms90
Sun Jul 12, 2015 11:38 pm
Forum: Estimation
Topic: Inference with GED distribution in GARCH models
Replies: 5
Views: 5394

Re: Inference with GED distribution in GARCH models

Alright. I have a large dataset, so asymptotics does seem valid. Thanks. I have a follow up question then. If the inference is done with asymptotic normal in any case, what is the effect of choosing error distributions? My data is very fat tailed, so GED does seem appropriate, however how does that ...
by Thomasms90
Sun Jul 12, 2015 1:12 pm
Forum: Estimation
Topic: Inference with GED distribution in GARCH models
Replies: 5
Views: 5394

Inference with GED distribution in GARCH models

Are the test statistics and p-values viable in a regular GARCH(1,1) model where the error term is specified as GED? Both with regards to the mean equation and variance equation. The output says "z statistic" which leads me to be in doubt as to whether inference is only reliable with normal...
by Thomasms90
Mon Jun 22, 2015 10:06 am
Forum: Estimation
Topic: Dynamic conditional correlation multivariate GARCH
Replies: 81
Views: 191095

Re: Dynamic conditional correlation multivariate GARCH

I can't quite understand how the likelihood function in the original code is specified. My algebra isn't that strong, so I'm trying to grasp it through looking at the articles and application code at the same time. Can anyone help me? dcc.append logl=(-1/2)*(2*log(2*pi)+log(detrRt)+(z1^2+z2^2-2*rho1...
by Thomasms90
Mon Jun 15, 2015 7:51 am
Forum: Estimation
Topic: Parameter restrictions in Switching Regressions
Replies: 5
Views: 5062

Re: Parameter restrictions in Switching Regressions

Alright. Was just in case there was a way to do it.

I'll manage without, no problem :-)
by Thomasms90
Fri Jun 12, 2015 5:52 am
Forum: Estimation
Topic: Parameter restrictions in Switching Regressions
Replies: 5
Views: 5062

Parameter restrictions in Switching Regressions

Is there any way to restrict coefficients in different regimes? I've only been able to restrict transition probabilities as it is. In my case I'm estimating a simple 3 regime markov process with only the dependant variable and a constant (switchreg y c - i.e. no exogenous regressors or AR terms). I ...
by Thomasms90
Wed Jun 10, 2015 3:40 pm
Forum: Estimation
Topic: (Co)Variance estimates with Gaussian kernel
Replies: 3
Views: 3645

Re: (Co)Variance estimates with Gaussian kernel

Thanks a bunch Glenn! Took a bit to figure out how it worked, but its a genious solution.

Reduced my complete execution time by a factor 115.
by Thomasms90
Wed Jun 10, 2015 8:30 am
Forum: Estimation
Topic: (Co)Variance estimates with Gaussian kernel
Replies: 3
Views: 3645

Re: (Co)Variance estimates with Gaussian kernel

I have tried to come up with a program to calculate the time varying standard deviation of a return series. Code is below. It seems to do the job pretty good, however it's incredibly inefficient. I have over 9000 observations leading to around 170 million iterations. Execution time is around 10-12mi...
by Thomasms90
Mon Jun 08, 2015 7:36 am
Forum: Estimation
Topic: (Co)Variance estimates with Gaussian kernel
Replies: 3
Views: 3645

(Co)Variance estimates with Gaussian kernel

Hi I'm interested in estimating time varying variances and covariances using a simple two sided gaussian/normal kernel function. However it is not one of the built-in options. Can anyone direct me towards how to specify the "User Kernel Vector" in order to do this if possible? Or pointers ...
by Thomasms90
Wed Mar 20, 2013 7:36 am
Forum: Data Manipulation
Topic: One month lag in daily data
Replies: 3
Views: 6863

Re: One month lag in daily data

That would be 28(29th) february. Basically the last day the previous month. We are using first days each month tho.. Still, if for instance we are looking at return 1. april, the 1 month lag would be 1. march. However if that's a sunday, we would want the day from 2. march.
by Thomasms90
Wed Mar 20, 2013 7:21 am
Forum: Data Manipulation
Topic: One month lag in daily data
Replies: 3
Views: 6863

One month lag in daily data

Hi, I have a dataset containing stock quotes and interest rates on a daily basis (5days/week). I wish to calculate the 1month returns to these, and thus need to be able to lag the values of variable by 1 month (and not 30 days or something similar). Is there any way to do this? I've tried applying a...

Go to advanced search