proc expand functionality

[annacole2408]

Hello, everyone. I am using the function convert daily/ observed = beginning

My daily data are a set of VIXCLS from FRED. When there is a value for the first day of the month, the output from the function above is the same as the original data. When there is not a value, however, it appears that SAS is somehow interpolating the data. I have tried several replications in MATLAB, using linear interpolation, spline interpolation, but have not been able to correctly replicate this function. Can someone please give insight into what the observed=OPTION function is doing when using the 'beginning' selection.

Thank you in advance

[ballardw]

I suggest including the entire Proc Expand code as options on many procedures may have interact with other options on other statements.

[Ksharp]

Firstly, Did you check the documentation of PROC EXPAND.
Secondly, Since PROC EXPAND is product under SAS/ETS. That would be better if you post it at Forecasting Forum:
https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/bd-p/forecasting_econometrics

[Kurt_Bremser]

I moved the question to the Forecasting community.

[sbxkoenk]

If the observations are equally spaced in time, and all the series are observed as beginning-of-period values, only the input and output data sets need to be specified. For example, the following statements interpolate any missing values in the numeric variables in the data set A using a cubic spline function, assuming that the observations are at equally spaced points in time:

proc expand data=a out=b;
run;

For more information, see the section "Missing Values".

The default is METHOD=SPLINE. 

Constraint specifications for METHOD=SPLINE can be specified.

The methods supported are SPLINE, JOIN, STEP, AGGREGATE, and NONE.

Ciao,
Koen