topic Re: How to Convert R code to SAS - ETS HP FILTER in SAS Forecasting and Econometrics

How to Convert R code to SAS - ETS HP FILTER

J111 — Sun, 23 Oct 2022 09:23:29 GMT

How can we convert this R code into SAS ETS ?

We have run this code on sashelp.air file - see results attached

-------------------------------------------------------------------------- The code -----------------------------------------------------------

#change the path

df<- read.csv("\download\AIR.csv")

lambda=26000

HP_FILTER_AIR= hpfilter(df$AIR,type=c("lambda"),drift=FALSE,freq=lambda)

df$cycle <- HP_FILTER_AIR$cycle

df$trend <- HP_FILTER_AIR$trend

---------------------------------------------------------------------------------------------------------------------------------------------------

Many thanks

Re: How to Convert R code to SAS - ETS HP FILTER

Ksharp — Sun, 23 Oct 2022 10:40:14 GMT

Calling @Rick_SAS

Check PROC TIMESERIES and its options:

OUTTREND= OUTSEASON=

P.S. check the examples in documentation,

Check PROC UCM

And I quote:

The UCMs are also called structural models in the time series literature. A
UCM decomposes the response series into components such as trend, seasonals, cycles, and the regression
effects due to predictor series.

Re: How to Convert R code to SAS - ETS HP FILTER

Rick_SAS — Sun, 23 Oct 2022 10:38:34 GMT

See this example in the documentation of the UCM procedure:

SAS Help Center: Trend Removal Using the Hodrick-Prescott Filter

Re: How to Convert R code to SAS - ETS HP FILTER

J111 — Sun, 23 Oct 2022 13:02:30 GMT

Thanks a lot - this code gives the wanted results !

proc expand data=sashelp.air out=fexp method=none ;
id date;
Convert air=hpt/transformout=(hp_t 26000) ;
Convert air=hpc/transformout=(hp_c 26000) ;
run ;