# Creating an earnings forecast with a partially completed year

Creating an earnings forecast can be challenging when the current year is not yet completed, and only partial data is available. In this blog post, we will explore two methods of creating an earnings forecast with a partially completed year: using only historical data, and using proration.

### Method 1: Using Only Historical Data

One way to create an earnings forecast with a partially completed year is to use only the historical data from past years, and ignore the partial data from the current year. This method assumes that the current year will follow the same trend and pattern as the past years, and that the partial data is not representative of the full year.

To use this method, you need to have several years of historical data. You can then calculate the compound annual growth rate (CAGR) of the earnings, which is the average percentage change in earnings from one year to the next. The formula for CAGR is:

CAGR = (Earnings [n] / Earnings [1])^1/(n – 1) – 1

where Earnings [n] and Earnings [1] are the earnings in the last and first years of history, respectively. Here n is the number of years in the earnings history you have available.

Once you have the CAGR,  use it to project the earnings for the current year and for a number of years into the future. This is done by multiplying the earnings in the last year by (1 + CAGR). For example, if the earnings in the last year were \$100,000, and the CAGR was 10%, then the projected earnings for the current year would be \$100,000 x (1 + 0.1) = \$110,000, and the projected earnings for the next year would be \$110,000 x (1 + 0.1) = \$121,000 and so on.

### Pros and cons of using historical data

This method looks simple and easy to use, but it has some limitations. It does not account for any changes or fluctuations in the current year, such as seasonality, market conditions, or unexpected events. It also assumes that the past performance is a reliable indicator of the future performance. This may not always be true. Hence, the method may not be very accurate or realistic, especially if the current year differs from the historic trend.

### Method 2: Using Proration

Another way to create an earnings forecast with a partially completed year is to use proration. This technique lets you adjust or allocate data based on a proportion or ratio. The method uses both the historical data and the partial data from the current year, and assumes that the partial data is representative of the full year.

To run this calculation, you need to have at least one year of historical data, and the partial data from the current year. Next figure out the prorated earnings for the current year, which is the earnings in the partial period multiplied by the ratio of the full period to the partial period. For example, assume the earnings in the first six months of the current year were \$50,000. The full year has 12 months. Then the prorated earnings for the current year would be \$50,000 x 2 = \$100,000.

Once you have the prorated earnings for the current year, you can apply the same CAGR formula as in the previous method to calculate the average annual growth rate of the earnings. To do so use the historical data and the prorated earnings. You can then use the CAGR to project the earnings for the next year, by multiplying the prorated earnings by (1 + CAGR). For example, let’s say that the prorated earnings for the current year were \$100,000, and the CAGR was 10%. Then the projected earnings for the next year would be \$100,000 x (1 + 0.1) = \$110,000.

### Pros and cons of proration

This method is more complex and requires more data than the previous method, but it has some advantages. It accounts for the changes and fluctuations in the current year, and does not rely solely on the past performance. It also allows for more flexibility and customization, as you can choose the partial period and the ratio that best suit your needs and preferences.

However, this method also has some limitations. It assumes that the partial data is representative of the full year, which may not always be true. It also depends on the accuracy and reliability of the partial data, which may be subject to errors or biases. Therefore, the results you get may not be reliable, particularly if the partial data is very different from the full year.