Forecasting Models

Introduction

Forecasting is a valuable way to help a company prepare and plan for the upcoming years. Forecasting however is not a fortune teller, it is designed as an educated guess of how much will be needed and utilized. It is a tool used to help budget and allocate money as accurately as possible.

Types of Forecasting Methods

Since there is not one right way of forecasting, there are many different types of forecasting. Each one is designed to help different organization and it is up to the company to choose the model which is best for them.

Qualitative

Some of the different types of forecasting methods include, Qualitative, “Subjective judgment based on estimates and opinions” (Chase, Jacobs, Aquilano, 2005, p. 514). These methods include talking with the people who are directly involved with sales to find out how a particular item or set of items are moving. A good example of Qualitative forecasting is Delphi. The Delphi model is similar to the panel consensus method. Both forecasting types require that a group of individuals are required to answer a questionnaire and/or share their ideas and opinions on a variety of topics. The argument against panel consensus is that everyone participating is aware of everyone else participating and knows how everyone else answers. This means that if the president of the company is in the group an entry-level person will know and that can cause the entry-level person to become shy and not fully express his or her opinions for fear of losing their job or being seen as not a team player. Delphi seeks to correct this apprehension. Delphi also utilizes a group of team members from the highest level to the lowest level. The difference is that the team members are handed questionnaires and asked to fill them out. They are then turned in and the mediator reads the answers. No one in the panel knows who voiced a certain opinion. This helps to get a more well rounded viewpoint without scaring those members who might worry about how their opinions will be perceived.

Time Series