cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
udo_sas
yesterday
  • 417 Posts
  • 27 Likes Given
  • 43 Solutions
  • 197 Likes Received

Re: Forecasting issue with proc varmax?

by SAS Employee in SAS Forecasting and Econometrics
‎03-01-2013 03:25 PM
‎03-01-2013 03:25 PM
Hello Ryan - Please excuse for delay - our developers have been able to replicate the problem which you are reporting. At this point in time I would suggest to open a track with Technical Support, who are already aware of this communication. Thanks! Udo ... View more

Re: Forecasting issue with proc varmax?

by SAS Employee in SAS Forecasting and Econometrics
‎02-25-2013 09:40 AM
‎02-25-2013 09:40 AM
Hello Ryan - Many thanks for your question about using VARMAX with differencing in the model. Our R&D team is currently investigating the question at hand - it would really help with you could provide us with a complete example - including data (even "faked" data will work. Another thought which crossed my mind was if a more simplistic approach might help addressing your problem at hand as well. If I understand correctly you would like to forecast "retvar" using "ratevar_shock100" as a input variable (assuming that there is cross-correlation between the two). In this case a UCM or ARIMAX approach might be applicable as well. Thanks! Udo ... View more

Re: Add index number around an event date

by SAS Employee in SAS Procedures
‎02-12-2013 05:22 PM
‎02-12-2013 05:22 PM
Anna - Not sure if I fully understand what you would like to see - I will give it another try (assuming that you have access to SAS/ETS software for the time being). I think what you are asking for is first of all how to fill gaps in your time id (date variable) and then to calculate relday, correct? If so, running the following code: *first fill gaps in timeid; proc timeseries data=want out=temp; id date interval=day accumulate=total; var abn_d; var rating_date / setmissing=prev; by eventid; run; *then calculate relday; data want; set temp; relday=date-rating_date; run; results in: If this is not what you have in mind a illustration of WANT would be useful. Thanks, Udo (aka Undo) ... View more

Re: Add index number around an event date

by SAS Employee in SAS Procedures
‎02-12-2013 02:22 PM
3 Likes
‎02-12-2013 02:22 PM
3 Likes
Hello - Does this address your "relday" question? data want;      set have;      relday=date-rating_date; run; Results in: Thanks, Udo ... View more

Re: help on SAS code...

by SAS Employee in SAS Forecasting and Econometrics
‎02-11-2013 06:40 PM
3 Likes
‎02-11-2013 06:40 PM
3 Likes
Hello - Never mind my earlier response - I guess you are running something like this (otherwise please do mind my earlier response :smileylaugh:): proc forecast data=have method=expo trend=2 outall outest=est lead=3;    id time;    var actual; run; You will find plenty of details of how the EXPO method of the FORECAST procedure works, here:  http://support.sas.com/documentation/cdl/en/etsug/63939/HTML/default/viewer.htm#etsug_forecast_sect023.htm (scroll down to EXPO Method). In particular you will find how to interpret the values of the OUTEST data set. May I also suggest to have a look at the ESM procedure instead - for your data set the code will look similar to this (I'm proposing to consider a DAMPTREND model, but you can change that of course): proc esm data=have out=outesm lead=3 outest=est2 plot=forecasts back=3;    id time interval=obs;    forecast actual / method=damptrend; run; which results in: Note that in this case the parameters listed in the OUTEST data set are easier to read: Thanks, Udo ... View more

Re: Forecast Studio 4.1 just hangs when I click on 'New' in New Projec...

by SAS Employee in SAS Forecasting and Econometrics
‎02-11-2013 06:10 PM
‎02-11-2013 06:10 PM
Hello Sunil - May I suggest to contact Technical Support on this issue? I'm sure my colleagues will be more than happy to track down what is causing the problem you are running into. Technical Support for SAS forecasting products is available using our online problem submission form. Thanks! Udo ... View more

Re: Handling weekly Seasonality in SAS Forecast Studio 4.1

by SAS Employee in SAS Forecasting and Econometrics
‎02-11-2013 11:41 AM
1 Like
‎02-11-2013 11:41 AM
1 Like
Hello Marcel - Please excuse for delay in responding - in order to test for seasonality Forecast Studio fits different autoregressive models behind the scenes and picks the most appropriate one based on AIC. For weekly data (seasonality of 52) the AIC penalizes models having too many parameters and a non-seasonal model will be preferred. As I have mentioned before - if you have knowledge about the seasonal behavior of your data, you are better off including seasonal models in a custom repository (for example BESTS). Happy to continue this discussion with you directly. Thanks! Udo ... View more

Re: How to check Decline trend and Upward trend Deviation?

by SAS Employee in SAS Forecasting and Econometrics
‎02-11-2013 11:26 AM
‎02-11-2013 11:26 AM
Hello - Yes, the MODEL statement of PROC UCM (SAS/ETS) allows you to specify a response variable and, optionally, the predictor or regressor variables for the UCM model. The predictors specified in the MODEL statement are assumed to have a linear and time-invariant relationship with the response. If your predictors have time-varying regression coefficients you can specify them separately in the RANDOMREG statement. Similarly, if the predictors have a nonlinear effect on the response variable you can specify them separately in the SPLINEREG statement. For more details on the UCM procedure you may want to check out: http://support.sas.com/documentation/cdl/en/etsug/63939/HTML/default/viewer.htm#etsug_ucm_sect001.htm Thanks! Udo ... View more

Re: help on SAS code...

by SAS Employee in SAS Forecasting and Econometrics
‎02-11-2013 08:40 AM
‎02-11-2013 08:40 AM
Hello Karteek - Would you mind to send your PROC FORECAST code as well? Thanks, Udo ... View more

Re: Handling weekly Seasonality in SAS Forecast Studio 4.1

by SAS Employee in SAS Forecasting and Econometrics
‎02-01-2013 04:16 PM
1 Like
‎02-01-2013 04:16 PM
1 Like
Hello Marcel - For your consideration: in "Forecast Settings" add "SASHELP.HPFDFLT.BESTS" as "models from an external list": This will force a couple of seasonal ESM to be run as part of your project. For details on BESTS please have a look at: http://support.sas.com/documentation/cdl/en/hpfug/63959/HTML/default/viewer.htm#hpfug_hpfdiag_sect012.htm Thanks, Udo ... View more

Re: Forecast Server Scenario analysis - Use time series models to pred...

by SAS Employee in SAS Forecasting and Econometrics
‎01-30-2013 08:52 AM
1 Like
‎01-30-2013 08:52 AM
1 Like
Hello Andreas - If you don't provide future values of independent variables, they are automatically forecast using one of the following smoothing models: simple double linear damped trend seasonal Winters method (additive and multiplicative) On the other hand, if you provide future values of your independent variables (similar to what you describe), these values will be taken into account. Thanks, Udo ... View more

Re: Seasonal Unit Root Tests not Produced in Forecast Server (and ETS)

by SAS Employee in SAS Forecasting and Econometrics
‎01-29-2013 08:56 AM
‎01-29-2013 08:56 AM
Hello - From documentation: In the Series View, the following options are now available from the Plots and Tables menus: Unit Root Tests provides Augmented Dickey-Fuller unit root tests for stationarity of the dependent series. The zero mean, single mean, and trend versions of the test are performed and unit root probabilities are shown at various lags using tables or bar charts. Seasonal Unit Root Tests are available if the series is seasonal with a seasonal cycle of up to 12 periods. Additional details can be found here: http://support.sas.com/documentation/cdl/en/hpfug/63959/HTML/default/viewer.htm#hpfug_hpfdiag_sect025.htm Thanks! Udo ... View more

Re: Dynamic Regression in Forecast Studio

by SAS Employee in SAS Forecasting and Econometrics
‎01-22-2013 10:37 AM
‎01-22-2013 10:37 AM
Hello Andreas - The TSFS dialog for custom models allows you to build "regression models with time series errors" which are also called transfer function models. The Error Model Options you specify are actually the p (AR) and q (MA) specifications you would enter in Forecast Studios ARIMA dialog (similar for P and Q in the seasonal ARIMA case). For example, specifying the following settings in the custom models dialog: can be replicated in Forecast Studio like this: Hope this helps. Thanks, Udo ... View more

Re: Forecasting in different time intervals with Forecast Server

by SAS Employee in SAS Forecasting and Econometrics
‎01-21-2013 08:51 AM
‎01-21-2013 08:51 AM
Hello - You might want to check out the HPFTEMPRECONCILE procedure, which reconciles high-frequency forecasts to the low-frequency forecasts in such a way that the accumulation of the reconciled high-frequency forecasts is equal to the low-frequency forecasts. For details check out: http://support.sas.com/documentation/cdl/en/hpfug/63959/HTML/default/viewer.htm#hpftemprecon_toc.htm Note that HPFTEMPRECONCILE is currently not supported in SAS Forecast Studio - so you would need to switch to code for the time being. In particular I would suggest to check out the SAS Forecast Batch Interface, which is documented in Chapter 11 of the SAS Forecast Server 12.1 Administrator’s Guide. Thanks! Udo ... View more

Re: Dynamic Regression in Forecast Studio

by SAS Employee in SAS Forecasting and Econometrics
‎01-16-2013 04:53 PM
3 Likes
‎01-16-2013 04:53 PM
3 Likes
Hello - As you probably already figured out, there are 2 dialogs related to ARIMA(X) in SAS Forecast Studio - between them you can do everything that TSFS can do (when it comes to ARIMAX) and you even have access to more features. When you create an Autoregressive Integrated Moving Average (ARIMA) model, you can specify the autoregressive and moving average polynomials of an ARIMA model. You can also include events and independent variables in the model. When creating a Subset (factored) ARIMA model, you can specify a general ARIMA model. You can specify the autoregressive and moving average polynomials of arbitrary complexity. You can also specify a general differencing order. The Subset ARIMA Model dialog box is similar to the ARIMA Model dialog box, except that it uses a more general specification of the ARIMA options. For more information about the ARIMA model and the Subset (factored) ARIMA model, see the HPFARIMASPEC procedure in the SAS High-Performance Forecasting: User's Guide. Thanks, Udo ... View more
Contact Me
Online Status
Offline
Date Last Visited
yesterday
Likes given to
Likes from
Quick links