Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: Forecast ARIMA
From
"Jeff" <[email protected]>
To
<[email protected]>
Subject
st: Forecast ARIMA
Date
Fri, 18 Jun 2010 13:35:22 -0700
Ladies, Gentlemen & Professors,
In Stata 10. I am attempting to teach myself time series in Stata.
I have an example of 72 monthly seasonal sales figures from one of the
books I am reading. (I think the example series came from "Forecasting,
Methods & Applications, 3rd Ed. by Makridakis, Wheelwright & Hyndman,
1998.) Using the Stata 10 drop down menu for the ARIMA process, I coded
in ARIMA(1,1,1) SARIMA (1,1,1,12). This seems to predict and graph the
model similarly to that which I produced in another program (Minitab,
which has limited capabilities allowing only univariate ARIMA
forecasting). However, going to "Dynamic Forecast" on the drop down
menu and plugging in the dependent variable Sales as the prefix, I don't
seem to be able to forecast (and consequently can't graph a forecast)
out for the next 24 months. I get the error message "fcast compute does
not work with arima". Obviously I am doing something wrong (probably a
whole lot wrong). Any ideas? Thank you in advance.
Also, any recommendation to a time series book which is more applied
than theoretical in nature, which illustrates Stata commands to a
variety of examples would be appreciated.
Jeffrey Wolpin
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/