# Economic Tools for Management Decision Making

Managerial decision-making draws on economic concepts as well as tools and techniques of analysis provided by decision sciences. The major categories of these tools and techniques are optimization, statistical estimation and forecasting. Most of these methodologies are technical. These methods are briefly explained below to illustrate how tools of decision sciences are used in managerial decision making.

## 1. Optimization

Optimization techniques are probably the most crucial to managerial decision making. Given that alternative courses of action are available, the manager attempts to produce the most optimal decision, consistent with stated managerial objectives. Thus, an optimization problem can be stated as maximizing an objective (called the objective function by mathematicians) subject to specified constraints. In determining the output level consistent with the maximum profit, the firm maximizes profits, constrained by cost and capacity considerations. While a manager does not resolve the optimization problem, he or she may make use of the results of mathematical analysis. In the profit maximization example, the profit maximizing condition requires that the firm select the production level at which marginal revenue equals marginal cost. This condition is obtained from an optimization model/technique. The techniques of optimization employed depend on the problem a manager is trying to solve.

## 2. Statistical Estimation

A number of statistical techniques are used to estimate economic variables of interest to a manager. In some cases, statistical estimation techniques employed are simple. In other cases, they are much more complex and advanced. Thus, manager may want to know the average price received by his competitors in the industry, as well as the standard deviation (a measure of variation across units) of the product price under consideration. In this case, the simple statistical concepts of mean (average) and standard deviation are used.

Estimating a relationship among variables requires a more advanced statistical technique. For example, a firm may desire to estimate its cost function i.e. the relationship between cost concept and the level of output. A firm may also wish to the demand function of its product that is the relationship between the demand for its product and factors that influence it. The estimates of costs and demand are usually based on data supplied by the firm. The statistical estimation technique employed is called regression analysis and is used to engender a mathematical model showing how a set of variables are related. This mathematical relationship can also be used to generate forecasts.

An example from the automobile industry is befitting for illustrating the forecasting method that employs simple regression analysis. Let us assume that a statistician has data on sales of American-made automobiles in the United States for the last 25 years. He or she has also determined that the sale of automobiles is related to the real disposable income of individuals. The statistician also has available the time series data (for the last 25 years) on real disposable income. Assume that the relationship between the time series on sales of American-made automobiles and the real disposable income of consumers is actually linear and it can thus be represented by a straight line. A rigorous mathematical technique is used to locate the straight line that most accurately represents the relationship between the time series on auto sales and disposable income.

## 3. Forecasting

It is a method or a technique to predict many future aspects of a business or any other operation. For example, a retailing firm that has been in business for the last 25 years may be interested in forecasting the likely sales volume for the coming year. Numerous forecasting techniques can be used to accomplish this goal. A forecasting technique, for example, can provide such a projection based on the experience of the firm during the last 25 years; that is, this forecasting technique bases the future forecast on the past data.