1
$\begingroup$

For one of my grad school classes I am working on a projection analysis. The data includes projected year-end balances. These balances are projected every month of the fiscal year (starting in July and ending in June) to see how much approximately is the Y-T-D amount going to be. I'd like to analyze the projected balances vs. the actual Y-T-D balance.
I'd like to find a way to estimate at which month we are comfortable at saying that this is approximately what the Y-T-D amount is going to be. I thought of using confidence interval and normal distribution, but I can't figure out how to approach this.

Thank you.

$\endgroup$
1
  • $\begingroup$ To make sure I understand: you are projecting the end-of-fiscal-year balance each month and obviously the projection becomes more accurate as the fiscal year progresses (and you have fewer months to project to the end of the year)? And you want to determine how far into the fiscal year you have to be to have a "firm" estimate of that year's ending balance? $\endgroup$
    – Wayne
    Nov 5, 2012 at 16:37

1 Answer 1

1
$\begingroup$

If you are projecting financial reports, look at this book.

If you are forecasting a time series, try an ARIMA or ETS model. The auto.arima(), ets(), bats() and tbats() functions in R's "forecast" package are pretty automatic.

$\endgroup$

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.