I have a data set of time series observations with a seasonal trend.
I have de-trended and de-seasonalized it using by first differencing and also taking the log of observations. However, when I want to fit an ARIMA model, It is almost impossible to have a proper fit.
This is my code and I have attached the data set.
proc import datafile="liquor.csv"
out=liquor
dbms=csv
replace;
getnames=no;
run;
data liquor (rename=(var1=sales));
set liquor;
t = _n_ ;
*The log transformation is predeferencing transformation to make constant seasonal variation;
LogSales=log(sales);
*These are three different differencing transformation to make the data stationary;
SALES1=logsales-lag(logsales);
SALES2=logsales-lag12(logsales);
SALES3=logsales-lag(logsales)-lag12(logsales)-lag13(logsales);
x=sales1-lag12(sales1);
run;
proc timeseries data=liquor
out=series
outtrend=trend
outseason=season print=seasons;
id date interval=month accumulate=avg;
var x;
run;
proc arima data=LIQUOR ;
identify var=x esacf ;
estimate p=2 q=3;
run;
Any suggestions and help is greatly appreciated.
I am closing this discussion due to inactivity.
I am closing this discussion due to inactivity.
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.