4
$\begingroup$

I am trying to forecast/predict the number of donations expected to be received at multiple locations across the country. I have and want to use the information I have. This information covers the previous 10 years of donations received up to and including the tenth day of the month in which the prediction is being made. The intent is to predict the number of donations to be received for the remainder of the current month. For example, I am on May 14, 2012. I want to predict the number of donations to be received from May 14, 2012 to the end of the month. I can use ten years' data up to and including data from May 10, 2012.

Since I want to try to use the forecasting method, how can this be done? I have the exact number, date and time of each donation received. This forecast must contain the weekends.

Please keep in mind that I am still trying to get used to this site and the way it operates. I have searched existing questions/answers and could not find something that would help me.

$\endgroup$
  • 3
    $\begingroup$ Are you trying to predict number or monetary amount of donations? $\endgroup$ – Deer Hunter Jan 2 '13 at 19:20
  • 2
    $\begingroup$ Asking in this way how to use "the forecasting method" seems to imply that there is a nice, unitary, cut-and-dried method. In actuality, if you really want to learn about forecasting you've got many weeks and probably months of study in front of you. And there are worse ways to spend one's time :-) $\endgroup$ – rolando2 Jan 2 '13 at 23:20
  • 1
    $\begingroup$ The approach was discussed at autobox.com/cms/index.php/blog/entry/… review that blog for the issue of making a month-end number. Specifically the idea is to take daily data & develop 1 equation that incorporates memory, day-of-the-week, day-of-the-month, week-of-the-year, month-of-the-year, the effect of events like holidays (pre and post), level shifts,local time trends.This equation can then be used to predict the remainder of the fiscal period and probability statements can then be made about achieving goals. $\endgroup$ – IrishStat Jan 3 '13 at 0:22
  • 2
    $\begingroup$ @DeerHunter trying to predict number of donations $\endgroup$ – user1834300 Jan 3 '13 at 13:56
  • 2
    $\begingroup$ Good comments by @DeerHunter and (at)rolando2. For improved search results if you want to forecast the "number of donations", try searching for "count models" and "non-negative integer valued ARIMA". A look at some literature might help give you some tried (and tested?) ideas, too. Difficult to suggest anything else without further information about the data. $\endgroup$ – Graeme Walsh Nov 30 '13 at 20:24
1
$\begingroup$

The approach was discussed at http://www.autobox.com/cms/index.php/blog/entry/what-business-intelligence-really-means-with-time-series-analysis review that blog for the issue of making a month-end number. Specifically the idea is to take daily data & develop 1 equation that incorporates memory, day-of-the-week, day-of-the-month, week-of-the-year, month-of-the-year, the effect of events like holidays (pre and post), level shifts,local time trends.This equation can then be used to predict the remainder of the fiscal period and probability statements can then be made about achieving goals.

You say you sent me the data , I don't recall ever receiving it. Can upu please contact me at my email address.

$\endgroup$
0
$\begingroup$

You say you sent me the data , I don't recall ever receiving it. Can upu please contact me at my email address.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.