*(second in a series)*

Today we discuss the various categories of forecasting methods that are available to businesses. Forecasting methods can be either objective (using quantitative approaches) or subjective (using more intuitive or qualitative approaches), depending on what data is available and the distance into the future for which a forecast is desired. Forecasting approaches will typically be more objective for nearer term forecasting horizons and for events where there is plenty of quantitative data available. More distant time periods, or events with a lack of historical quantitative data will often call for more subjective approaches. We will discuss these two classes of forecasting methods, and the categories within each.

**Objective Forecasting Approaches**

Objective forecasting approaches are quantitative in nature and lend themselves well to an abundance of data. There are three categories of objective forecasting methods: time series, causal/econometric, and artificial intelligence. AI approaches are outside my experience, so I won’t be covering them in this series, but mention them as another alternative, in case you wish to investigate them on your own.

*Time Series Methods *

Time series methods attempt to estimate future outcomes on the basis of historical data. In many cases, prior sales of a product can be a good predictor of upcoming sales because of prior period marketing efforts, repeat business, brand awareness, and other factors. When an analyst employs time series methods, he/she is assuming that the future will continue to look like the past. In rapidly changing industries or environments, time series forecasts are not ideal, and may be useless.

Because time series data are historical, they exhibit four components that emerge over time: trend, seasonal, cyclical, and random (or irregular). Before any forecasting is done on time series data, the data must be adjusted for each of these components. Decomposing time series data will be discussed later in this series.

The most common time series methods include moving average (both straight and weighted), exponential smoothing, and regression analysis. Each of these approaches will be discussed later in the series.

*Causal/Econometric Methods*

Causal or econometric forecasting methods attempt to predict outcomes based on changes in factors that are known – or believed – to impact those outcomes. For example, temperature may be used to forecast sales of ice cream; advertising expenditures may be used to predict sales; or the unemployment rate might be used to forecast the incidence of crime in a neighborhood. It is important to note, however, that just because a model finds two events that are correlated (e.g., occur together), it does not necessarily mean that one event has caused the other.

Regression analysis also falls under the causal/econometric umbrella, as it can be used to predict an outcome based on changes in other factors (e.g., SAT score may be used to measure likelihood of being accepted to a college). Econometric forecasting methods include Autoregressive Moving Average (ARMA) and Autoregressive Integrated Moving Average (ARIMA) models. ARIMA was previously known as Box-Jenkins. ARMA and ARIMA models are used in certain cases, but most of the time are unnecessary. Although these two methods won’t be covered in much depth later in the series, there will be a brief description of them and when they are needed.

**Subjective Forecasting Approaches**

Subjective forecasts are more qualitative. These approaches rely most heavily on judgment and educated guesses, since there is little data available for forecasting. This is especially the case in long-range forecasting. It’s easy to forecast next week’s sales of ice cream – and possibly even of individual flavors, since you’ll likely have months or years of past weekly ice cream sales data. However, if you’re trying to get an idea of what ice cream consumption or flavor preferences will be 10 years from now, quantitative approaches will be of little use. Changes in tastes, technology, and political, economic, and social factors occur and can dramatically alter the course of trends. Hence, the opinion of subject matter experts is often called upon. There is essentially only one category of subjective forecast approaches – and it is rightly called “Judgmental” forecasts.

*Judgmental Methods*

Judgmental forecasting methods rely much on expert opinion and educated guesses. But just because they have little quantitative or objective basis doesn’t mean they should be dismissed or not measured for accuracy. The most common types of of judgmental forecasting methods are composite forecasts, extrapolation, surveys, Delphi method, scenario writing, and simulation. Each of these methods will be discussed in detail later in the series.

*Introducing “Forecast Fridays” – ON THURSDAYS!!!*

Beginning with part 3, which will discuss moving average forecasts, the forecasting series will begin posting weekly so that the remaining days of the week can still be devoted to other topics in the marketing research and analytics field. The weekly post will be called “Forecast Friday.” However, it will be posted *every Thursday! *Why? Find out in tomorrow’s post!

Tags: ARIMA, ARMA, artificial intelligence, Autoregressive Integrated Moving Average Models, Autoregressive Moving Average models, Box-Jenkins, causal modeling, Delphi method, econometric forecasting, econometrics, exponential smoothing, extrapolation, Forecast Friday, Forecasting, judgmental forecasting, moving average, predictive modeling, regression, scenario building, scenario writing, simulation, Surveys, time series analysis

## Leave a Reply