Hello - Both ARIMA and UCM will not only give you forecasted values, but much more. A nice feature of UCM is that it can "decompose" forecasts into their components - which will allow you to do the "what-if" type of scenario you are talking about. Below you will find an example - hope this helps. Thanks, Udo *first create future values for the masonry variable of sashelp.workers; proc esm data=sashelp.workers out=masonry plot=forecasts; id date interval=month; forecast masonry / method=winters; run; *merge the 2 new data sets - note that for masonry it features forecasted values, which you can change later to assess the impact; data total; merge sashelp.workers(drop=masonry) masonry; by date; run; *run UCM with mansonry, irregular and level (which is also called trend in UCM) as compoents; *check out the outfor data set, which features the components of the forecasts; *outest features the parameter estimates; proc ucm data=total plots=smooth(decomp);    id date interval=month;    model electric=masonry;    irregular;    level;    estimate outest=outest;    forecast lead=12 outfor=outfor; run; ... View more

Hello - Given your description you may want to check out either ARIMA or UCM of SAS/ETS software. Thanks, Udo ... View more

Hello - proc timeseries data=sashelp.air plot=(series periodogram spectrum) out=_null_;    var air;    id date interval= month;    spectra freq period p s / adjmean=yes bartlett domain=time; run; seems to run fine - which problem are you running into? Thanks, Udo ... View more

Hello - Yes - if you are willing to share your data just attach it to a post. Thanks, Udo ... View more

Allie - Would it be possible to share your data as well? Thanks, Udo ... View more

Hi Akilees - My suggestion would be to start with a univariate model and try to add complexity like input variables later. Please have a look at PROC ESM as a start - you will need to make sure that you data is in time series format in order to use that (or any other time series forecasting) procedure. Check out: http://support.sas.com/documentation/cdl/en/etsug/66100/HTML/default/viewer.htm#etsug_esm_gettingstarted.htm as a starting point - you will find lots of information about time series data here: http://support.sas.com/documentation/cdl/en/etsug/66100/HTML/default/viewer.htm#etsug_tsdata_toc.htm The way I see it you will need to do something like: PROC ESM ...; id week_ending interval=week; forecast sizeAvol SizeBvol; run; Note that this code will run a simple ESM by default, so you may want to look at the model option of the FORECAST statement for more advanced models. For more complex models, such as ARIMAX or UCM, which allow you to take independent variables into account please refer to either PROC ARIMA or PROC UCM. All of these procedures are features in SAS/ETS software. Thanks, Udo ... View more

Yes - the first link should download the whitepaper - here is link: http://www.sas.com/reg/wp/corp/3478 Thanks, Udo ... View more

Hello - It seems to me that the optimization solver did not converge. In my opinion you should either change the nonlinear optimization technique or increase the number of iterations for your current solver. For details please refer to http://support.sas.com/documentation/cdl/en/etsug/66100/HTML/default/viewer.htm#etsug_varmax_syntax17.htm and  http://support.sas.com/documentation/cdl/en/etsug/66100/HTML/default/viewer.htm#etsug_nlomet_toc.htm Thanks, Udo ... View more

Hello - From: Large-Scale Automatic Forecasting Using Inputs and Calendar Events: "Causal time series models are used to forecast time series data that is influenced by causal factors. Input variables (regressor or predictor variables) and calendar events (indicator, dummy or intervention variables) are examples of causal factors. These independent (exogenous) time series causally influence the dependent (response, endogenous) time series, and therefore can aid the forecasting of the dependent time series. Examples of causal time series models are autoregressive integrated moving averages with exogenous inputs (ARIMAX), which are also known as transfer function models, and dynamic regression models and unobserved component models (UCM), which are also known as state-space models and structural time series models." SAS/ETS will give you access to these models in the following procedures: AUTOREG, ARIMA, UCM, or SSM. SAS Forecast Server (and SAS Forecasting for Desktop) will automatically build causal models for you (for ARIMAX and UCM) considering both independent variables and events. Currently the ESM procedure does not allow for independent variables. Thanks, Udo ... View more

Hello - This is a very general question indeed - but I'll take a shot at it, maybe you could provide some background for your question. For forecasting modeling challenges exponential smoothing models will do a good job typically. Of course I'm talking about ESM with optimized parameters, which are available both in SAS/ETS (in PROC ESM) or in SAS Forecast Server. Of course they can fall short as well, in particular if there is too little history available or if your time series patterns are sporadic or intermittent. Is this helpful at all? Thanks Udo ... View more

Hello Jean - Not sure if this is what you are wondering about, but it might be as easy as adding an ODS output statement to your code. Thanks, Udo Example: data air;    input ozone @@;    label ozone  = 'Ozone Concentration'          x1     = 'Intervention for post 1960 period'          summer = 'Summer Months Intervention'          winter = 'Winter Months Intervention';    date = intnx( 'month', '31dec1954'd, _n_ );    format date monyy.;    month = month( date );    year = year( date );    x1 = year >= 1960;    summer = ( 5 < month < 11 ) * ( year > 1965 );    winter = ( year > 1965 ) - summer; datalines; 2.7  2.0  3.6  5.0  6.5  6.1  5.9  5.0  6.4  7.4  8.2  3.9 4.1  4.5  5.5  3.8  4.8  5.6  6.3  5.9  8.7  5.3  5.7  5.7 3.0  3.4  4.9  4.5  4.0  5.7  6.3  7.1  8.0  5.2  5.0  4.7 3.7  3.1  2.5  4.0  4.1  4.6  4.4  4.2  5.1  4.6  4.4  4.0 2.9  2.4  4.7  5.1  4.0  7.5  7.7  6.3  5.3  5.7  4.8  2.7 1.7  2.0  3.4  4.0  4.3  5.0  5.5  5.0  5.4  3.8  2.4  2.0 2.2  2.5  2.6  3.3  2.9  4.3  4.2  4.2  3.9  3.9  2.5  2.2 2.4  1.9  2.1  4.5  3.3  3.4  4.1  5.7  4.8  5.0  2.8  2.9 1.7  3.2  2.7  3.0  3.4  3.8  5.0  4.8  4.9  3.5  2.5  2.4 1.6  2.3  2.5  3.1  3.5  4.5  5.7  5.0  4.6  4.8  2.1  1.4 2.1  2.9  2.7  4.2  3.9  4.1  4.6  5.8  4.4  6.1  3.5  1.9 1.8  1.9  3.7  4.4  3.8  5.6  5.7  5.1  5.6  4.8  2.5  1.5 1.8  2.5  2.6  1.8  3.7  3.7  4.9  5.1  3.7  5.4  3.0  1.8 2.1  2.6  2.8  3.2  3.5  3.5  4.9  4.2  4.7  3.7  3.2  1.8 2.0  1.7  2.8  3.2  4.4  3.4  3.9  5.5  3.8  3.2  2.3  2.2 1.3  2.3  2.7  3.3  3.7  3.0  3.8  4.7  4.6  2.9  1.7  1.3 1.8  2.0  2.2  3.0  2.4  3.5  3.5  3.3  2.7  2.5  1.6  1.2 1.5  2.0  3.1  3.0  3.5  3.4  4.0  3.8  3.1  2.1  1.6  1.3 .    .    .    .    .    .    .    .    .    .    .    . ; ods OUTPUT ParameterEstimates=work.estimates; proc arima data=air;    /* Identify and seasonally difference ozone series */    identify var=ozone(12)    crosscorr=( x1(12) summer winter );    /* Fit a multiple regression with a seasonal MA model */    /*     by the maximum likelihood method               */    estimate q=(1)(12) input=( x1 summer winter )    noconstant method=ml;    /* Forecast */    forecast  lead=12 id=date interval=month; run;quit; proc print data=work.estimates;run; ... View more

Hello - You can find an example here: http://support.sas.com/documentation/cdl/en/etsug/66100/HTML/default/viewer.htm#etsug_tsdata_sect004.htm and more information here: http://support.sas.com/documentation/cdl/en/etsug/66100/HTML/default/viewer.htm#etsug_tsdata_sect004.htm In order to introduce hierarchies like type, state, retailer you will need to use BY variables. See:  http://support.sas.com/documentation/cdl/en/etsug/66100/HTML/default/viewer.htm#etsug_tsdata_sect024.htm If you check out the SASHELP.PRICEDATA sample data set, you will get an impression of the structure needed. Thanks, Udo ... View more