SAS：用所有数据绘制预测数据

Question

我使用 proc ARIMA 来获取我的数据集下一年的预测，并有一个包含这些预测值的输出数据集。

我将原始数据的对数用于预测，因此我需要通过对预测值取指数来获得真实值。

我想要一个图表来显示数据的预测方式以及与原始数据相比的外观，但我不知道如何执行此操作。

代码：

* Open the file;
data intel_stock;
infile 'path' dlm=',' firstobs=2;   
input Date anydtdte10. Volume;
format Date date10.;
Timeref=_n_;
logvolume = log(Volume);
run;

* Plot the data ;
proc sgplot data=intel_stock;
series x=Timeref y=Volume/markers;  
xaxis values=(1 to 5000 by 1);  
run;
* Variation seems to increase greatly over time, hence we take the log of volume ;

proc sgplot data=intel_stock;
series x=Timeref y=logvolume/markers;   
xaxis values=(1 to 5000 by 1);  
run;
* Plot shows a good amount of variance removed ;

* selecting an ARIMA model ;
proc arima data=intel_stock;
identify var= logvolume(1); * first difference was taken to make the data stationary ;
estimate p = 2 q = 2 ;
forecast lead=12 interval=month id=Date out=forecast;
run;
* ARIMA(2,1,2) model was used;

Answer 1

使用Proc ARIMA 选项PLOTS=ALL。

例子：

%if not %sysfunc(cexist(work.sasmacr.yahoo_stock_quotes.macro)) %then %do;
filename source url "https://www.devenezia.com/downloads/sas/macros/download.php?file=stock_quotes.sas";
%include source;
filename source;
%end;

%if not %sysfunc(exist(work.intc)) %then %do;
%yahoo_stock_quotes(symbol=INTC,start=11/01/2016)
data INTC;
  set INTC;
  Timeref=_n_;
  logvolume = log(Volume);
run;
%end;

ods html file='intc.html';

proc sgplot data=INTC;
scatter x=Timeref y=Volume ;
run;
proc sgplot data=INTC;
scatter x=Timeref y=logvolume ;
run;

proc arima data=INTC plots=all;
title "INTC - ARIMA(2,1,2)";
identify var=logvolume(1); * first difference was taken to make the data stationary ;
estimate p = 2 q = 2 ;
forecast lead=12 interval=month id=Date out=forecast;
run;
quit;

ods html close;

Answer 2

* Removes logarithm from volume, forecast, upper and lower 95% CIs;
data intel_forecast;
set forecast;
Volume = exp(logvolume);
l95 = exp(l95);
u95 = exp(u95);
forecast = exp(forecast + std*std/2);
run;

* plots forecast with the rest of the data ;
proc sgplot data=intel_forecast;
where date >= '1JAN18'D;
band Upper=u95 Lower=l95 x=Date
/ legendLabel="95% Confidence Limits" ;
scatter x=Date y=Volume;
series x=Date y=forecast
/ legendlabel="Forecast of Volume for the next 5 years";
run;