# Regression Equations of Real World Value

Please help me with these questions:

1. What is the real world value of the y-intercept and slope in a regression equation?

2. Other than statistics problems, why should I care?

3. I know how to calculate, but what does it really mean when you apply it in a business setting?

4. When you use them in forecasting, how accurate can it really be? After all, there are so many factors (other than the regression equation) to consider?

https://brainmass.com/statistics/regression-analysis/regression-equations-real-world-value-41095

#### Solution Preview

1. What is the real world value of the y-intercept and slope in a regression equation?

It sounds like you skeptical of making future predictions based on statistical methods (i.e., regression equation).

2. Other than statistics problems, why should I care?

Keep reading...

3. I know how to calculate, but what does it really mean when you apply it in a business setting?

It actually means a lot. Business planning, for example, depends on see what factors correlate in order to predict (of course considering the uncertainty involved with any prediction usually set at .05 meaning that 5 times out of 100, the prediction will miss the mark). See many examples listed below.

4. When you use them in forecasting, how accurate can it really be? After all there are so many factors (other than the regression equation) to consider?

True, but regression equation is one method used in forecasting. No one method will provide the whole picture, of course, but it one of the mathematical techniques needed for forecasting, necessary in the area of business planning. It is important for prepare graphs of linear and quadratic equations for reports. Other things are also important, such as calculate the future value of regular savings (sinking funds) or find the savings given the future value, if necessary, using the sum of a geometric progression. For example, calculating the NPV of a project and use this to decide (forecasting) whether a project should be undertaken, or to choose between mutually exclusive projects.

Choosing and using business forecasting

Every forecast you make is a "conditional" statement of what will happen in the future. The forecast depends on what also happens to the surrounding situation--the effect of coffee prices on tea sales depends upon consumer's expectations about future coffee prices, whether there is an excess supply of tea at current prices, and the exchange rate between the United States and tea/coffee producing countries. Every forecast is, then, limited by the "conditioning events" which surround the event to be forecasted. Assumptions of Every Forecast

While instinct and estimates will always have their place in business (some of us are better than others at following hunches) managers are lately turning toward systematic and objective forms of forecasting. An objective forecast is simply one, which results from the forecaster using a model to make the forecast. A model, we will see, is just a compact statement of the way you think things work. Any model used for forecasting today is based on three simple assumptions:

* Future occurrences are based, at least in part, on presently observable events.

* Things will behave in the future much as they have in the past.

* The relationships that have occurred in the past can be discovered by study and observation. Systematic Forecasting

Systematic forecasting assumes that we can observe the underlying relationships that have occurred in the past by blocking out much of reality and building abstractions (models) which take into account only those things we feel are of prime importance in predicting something. It is no wonder that forecasters are accused of being simplistic and unrealistic--they are! In fact, to be simplistic is the only way to make any sense out of the complex relationships we face in the real world.

Forecasts may consist of predicting amounts, probabilities, or the timing of an event. We may all feel certain, for instance, that man will ultimately fly to Mars. If we were to predict when man would fly to Mars that would be a "timing forecast." If we were to predict the probability that the Dallas Cowboys would make it to the Superbowl, that would be a probability forecast. In this article we will deal with neither timing nor probability forecasts.

Here we will concentrate on predicting quantities. This is, by far, the most common form of business forecasting. If your company wants a sales forecast for next quarter, that is a quantity forecast. A cash flow forecast or an inventory forecast would also be a quantity forecast.

The systematic approach to forecasting may take the form of a carefully constructed model which the forecaster builds to mimic a real-world situation in which the assumptions are set up in strict mathematical form; this method is called econometrics. Or the forecaster's approach may be much less rigorous and much more dependent upon intuition and whatever data are readily available. Either approach to forecasting requires essentially the same statistical tools. Widely Used Forecasting Techniques

While many techniques for forecasting require a thorough study of economics and statistics, we shall present a set of elementary forecasting techniques most of which are available in most of the software packages listed in the accompanying comparison chart. We purposely overlook some forecasting techniques (such as using leading indicators or surveys of economic intentions) which, while quite useful, do not directly incorporate computer models.

The techniques we will cover include:

* Linear Regression--a method for using one variable to predict a second variable.

* Multiple Linear Regression--a method for using more than a single variable to predict another variable.

* Time Series Analysis--a way of studying the movement of a variable over time in order to predict its future values. Simple Linear Regression Model

A problem encountered by almost every manager is how to predict the value of some variable when the forecast variable is assumed to be dependent upon (or caused by) another variable.

For example, assume a carpet manufacturer finds that the number of residential building permits issued in a given quarter is strongly related to the company's carpet sales in the next quarter (this example is taken from the Graph 'n Calc ...

#### Solution Summary

By responding to the questions, this solution addresses aspects of regression analysis and the regression equation including concrete real-world business examples and its use in forecasting.