Forecasting

  • -

Forecasting

Category : Articles

???index and Forecasting???

QRB 501
University of Phoenix Online

Trevor Sheppard
July 5, 2010

Introduction
A class assignment was given to forecast for a given 5th year. There are so many different ways that people could forecast, because forecasting is nothing other than an educated guess. They may not even be correct was that final year is over, but it helps to try to see what is going to happen in the future. Uses for forecasting include creating sales goals for future use and to decide on stocks. Many companies use forecasting on a regular basis, but it is just like the weather on the news. On some days, they will say 0% chance of rain, and there will be rain. On other days they will say 100% chance of rain, and it will not rain a drop. Here is just one method of forecasting using trend lines.
Indexes
First, I got the regular index because that was part of the assignment. The index is found by using the equation P1/P0 (100) = Index. P1 is the current year. P0 is the base year, or in this case it is year 1, month 1. I also used a seasonal index. It is where you calculate things based on a moving average. The seasonal index is calculated like this: Seasonal index = seasonal average / grand average * 100. Instead of taking one month at a time it takes the quarterly average and keeping moving. For instance, the seasonal would the average for months 1, 2, and 3, but for month 2 it will be months 2, 3, and 4. This method helps prevent getting unreal forecast, because one month may have an unusually good month.
Trend Lines
Trend lines were used to find the linear equation that solves the demand for any given month. The trend line is the basis for getting the forecast for the 5th year. It assumes that everything is traveling in a straight line, and finds the best path. That is how the equation is produced. The equation for this one is on the graph for linear. Once the equation is gotten, it is then put into the cells that need to be forecasted. The equation for this example is adjusted forecast = (404.16*x) + 34426, where x is the seasonal index. Once the seasonal index is plugged it, it gives the plot on the linear line for that month, but everything is not perfect. Then the adjusted forecasted is taken, and multiplied by the seasonal index to get the final answer. That tells how are far off of the trend line that the number should fall.
Conclusion
Trend lines show the linear path that a line that is not straight should take if the conditions were perfect, but real life is not perfect. The numbers have to be adjusted to forecast real life numbers, because the forecast should be as close to real life as possible. Using statically features allows for the forecast to become more realistic than just doing a rough guess. The process may seem long, but it is necessary to make sure that the prediction is very possible and realistic.