09-01-2015
Xman
Calcite | Level 5
Member since
01-19-2013
- 6 Posts
- 0 Likes Given
- 0 Solutions
- 0 Likes Received
-
Latest posts by Xman
Subject Views Posted 2623 12-02-2014 03:17 AM 4000 12-02-2014 03:14 AM 4683 12-01-2014 02:37 AM 3043 11-28-2014 06:06 AM 1620 01-19-2013 04:55 AM -
Activity Feed for Xman
- Posted Re: Holdout MAPEs in SAS PROC ARIMA and SAS Forecast Studio don't match on SAS Forecasting and Econometrics. 12-02-2014 03:17 AM
- Posted Re: Holdouts in Proc ARIMA on SAS Forecasting and Econometrics. 12-02-2014 03:14 AM
- Posted Holdouts in Proc ARIMA on SAS Forecasting and Econometrics. 12-01-2014 02:37 AM
- Posted Holdout MAPEs in SAS PROC ARIMA and SAS Forecast Studio don't match on SAS Forecasting and Econometrics. 11-28-2014 06:06 AM
- Posted Re: Classification of observations after Ward's minimum variance on Statistical Procedures. 01-19-2013 04:55 AM
12-02-2014
03:17 AM
HI Udo - Thank you for your helpful comments! I guess I have covered my questions on the back= option in another post so will not duplicate it here. The link you provided has only the abstract of the talk... could you please guide me to the full presentation? That would be very helpful! Many Thanks again! Regards, Xe
... View more
12-02-2014
03:14 AM
Many Thanks for your helpful comments, Udo! But I still am not sure I understand what the back= option does in PROC ARIMA; From the manual: "specifies the number of observations before the end of the data where the multistep forecasts are to begin. " My question is: is the PROC using the data, including the data specified by the back= option to estimate the parameters? The manual only says it begins multistep forecasts from a prior period, nothing about the data being used or not used. Also, my experiments with various values of Back= gave the same parameter values, suggesting that the proc uses up all the data... In light of this, I am not sure I understand your comment on Back= specifying the out of sample part. Could you please elaborate? As an aside, The Proc UCM manual is quite clear in its description of the Back= option BACK=integer specifies the holdout sample for the evaluation of the forecasting performance of the model. For example, BACK=10 results in treating the last 10 observed values of the response series as unobserved. A post-sample-prediction-analysis table is produced for comparing the predicted values with the actual values in the holdout period. The default is BACK=0. Is this the case with PRoc ARIMA too? Many Thanks in advance! X
... View more
12-01-2014
02:37 AM
HI, My problem: I need to build many ARIMA models and choose the ones with low MAPEs on Holdout samples, sort of like what Forecast Studio does. I am running the following ARIMA statements in a loop, trying out various values of p, q and inputs: Y is my response series, X is a input series I use to explain the Ys. proc ARIMA data=Input_dataset; Identify Var=Y crosscorr=X; Estimate p=1 q=1 input =X; Forecast id=DAte interval=months out=REsults Lead=12 back=12 nooutall; quit; proc sql; select mean(abs(residual)/Y)*100 as MAPE, mean((residual)/Y)*100 as MPE from results; quit; My understanding is that the Back=12 statement tells SAS not to use the final 12 datapoints for estimating the parameters and the lead=12 forecasts the values for these final 12 datapoint. Then the nooutall statement forces only the last 12 datapoints into the REsults dataset. The final Proc SQL statements uses these 12 actual and forecasted values to estimate the holdout MAPES. So I my questions are: Is the programming structure above creating a holdout sample? If Not, what should I be doing to get what I am trying to accomplish ?
... View more
11-28-2014
06:06 AM
I have a Time Series (ARIMA) model in SAS, modelled using proc ARIMA which I am trying to replicate in SAS Forecasting Studio. What I see is that The parameter estimates in both are very similar (which is surprising, since my colleagues haven't had such luck, but I digress), but the MAPEs calculated with a Holdout of 7 months does not match. I would understand the parameters not matching, with the numerical optimization algorithms giving different results with different initial values, but the MAPEs not matching is very puzzling. Has anyone else here come across this, or has any idea what might be the case? DO you need any more information to address this issue? Thanks, V
... View more
01-19-2013
04:55 AM
Check out Proc Tree. You will need to save the cluster output to a dataset that Proc Tree would use. Check out the documentation for both.
... View more