I am trying to graphically determine an estimated projected date in the
future that an account will reach a certain balance, and need a formula to
resolve this.

Because the amounts being credited are random in date and amount, I need to
simplify things by assuming that the next amount credited will be a mean
average of all credits to the account spread over the time since the first
payment. Since the data is organised one row per day, each day should show
the average of all payments made so far since day one, and the same each day
in the future, so we can create a graph of projected balance on any given
future date.

This will change over time, depending on the amount and frequency of
payment. In column A I have dates, in B a balance, which is calculated as
the sum of all credits so far and column C holds payments to the account.

Example

1 Jan 10------0------0
2 Jan 10------5------5
3 Jan 10------5------0
4 Jan 10-----15-----10
5 Jan 10-----15------0
6 Jan 10-----22------7
7 Jan 10-----28------6
......

Column D will hold the calculation for the projection, this will be graphed
and the resulting line will be used to calculate the projected date when it
intersects with the projected amount.

Row 1 has column headings and in D2 I have =AVERAGE($C$2:C2) and then
=B2+AVERAGE($C$2:C3), =B3+AVERAGE($C$2:C4) ...etc

But currently my problem is that although the formula calculates the
average, it hasn't (as yet) got figures for amounts credited on future
dates, since the payments are irregular, the average decreases going down
the column (since an average spread over a greater number of days will
decrease), where I would like it to remain the same until the day actually
passes and nothing has been credited.

Any ideas?