Contrast qualitative, historical projection, and causal forecasting models. What are the strengths and weaknesses of each of these models?
Three demand forecasting models are qualitative, historical projection methods and causal methods. In the qualitative methods, experts use judgment, prior experiences, surveys, or comparative techniques to provide quantitative estimates about the future. A team of experts provide insights about the forecast or some predictive analysis of the demand forecast. This method is usually adopted when there is not much past data available for the demand. When historical data is available, and the trend and seasonable variations are stable and well defined, then we go for historical projection methods. The basic premise is that the future time pattern will be a replication of the past. The quantitative nature of the historical projection methods encourages the use of statistical models. These models work because of the inherent stability of the historical projection methods in the short run. Cause-and-effect is based on the principle that there are one or more factors are related to demand and that the relationship between cause and effect can be used to estimate future demand. Examples of cause-and-effect forecasting include simple and multiple regressions. In simple regression, demand is dependent on only one variable, whereas in multiple regressions, demand is dependent on two or more variables. Judgmental methods can be sometimes preferred over other techniques as it has an ability to incorporate unusual events, and if there is difficulty of obtaining the data necessary for quantitative techniques.