Date:         Tue, 19 Oct 2010 00:26:00 +0100
Reply-To:     "adel F." <adel_tangi@YAHOO.FR>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "adel F." <adel_tangi@YAHOO.FR>
Subject:      proc arima
Content-Type: text/plain; charset=utf-8

Hi all,

I would be very thankful if you could comment on my pgm below. I have followed 
example 7.2 (International airline passengers) in chapter 7 (the arima 
procedure).

My problem is the following , I have a dataset on people being recruited for a 
trial from January 08 to August 10. We aim to recruit 4500. My objective by 
using proc arima is to predict the time I will need to acheive the target of 
4500 by July 2012. 


I have used time series analysis, I would be thankful if you could advise 
whether the approach I have taken is correct or not, or whether the model I have 
considered is correct or not.

Below is the dataset and the modelling

Thanks a lot 

Adel

data mydataset;
retain time 0;
   input date:monyy5. numpeople @@;
   time = time + 1;
   lag_people = LAG(numpeople);
   xlog=log(numpeople);
   format date monyy5.;
datalines;
Jan08    38
Feb08    33
Mar08    38
Apr08    37
May08    49
Jun08    48
Jul08    66
Aug08    47
Sep08    70
Oct08    66
Nov08    72
Dec08    41
Jan09    75
Feb09    89
Mar09    82
Apr09    70
May09    55
Jun09    76
Jul09    79
Aug09    49
Sep09    82
Oct09    85
Nov09    81
Dec09    64
Jan10    63
Feb10    98
Mar10    95
Apr10    75
May10    89
Jun10    96
Jul10    103
Aug10    69
;
run;
symbol i=join v=dot;
proc gplot data=mydataset;
plot numpeople* date =1;
run;
proc arima data=mydataset;
identify var=xlog(1,12) nlag=15;
estimate q=(1)(12)  noconstant method=uls;
run;
forecast out=b lead=24 id=date interval=month noprint;
run;
data c;
set b;
x=exp(xlog);
forecast=exp(forecast +std*std/2);
l95=exp(l95);
u95=exp(u95);
run;
symbol1 i=none v=star;
symbol2 i=join v=circle;
symbol3 i=join v=none l=3;
proc gplot data=c;
where date >= '1JAN08'd;
plot x*date = 1 forecast * date = 2
l95 * date = 3 u95 * date = 3 /
overlay haxis= '1JAN08'd to '1AUG12'd by year;
run;