TABLE OF CONTENTS Forecasting is used to predict future events based upon estimates using historical data as a Baseline Comparative System (See Appendix D). Forecasting tools can be used to predict the outcome of technological progress upon a system or component, the effects of economic changes, and inventory demand rates. These time series forecasting methods use related data points corresponding to periods of time to attempt to predict a future occurrence.
The correlation coefficient determines how well the data fits the straight line. It is represented by the letter "r" and its values range from +1.0 (perfect positive correlation) to -1.0 (perfect negative correlation). A coefficient of zero (0) means that the data does not follow the line at all. A coefficient of one (1) means that all of the data points lie on the line.
Example:
A freight forwarder finds that an increasing amount of railroad boxcars are shipped empty from West to East every month. Determine the degree of correlation between the returning railroad boxcars using the monthly report shown next.
| Sample Data | |||||
|---|---|---|---|---|---|
| January | February | March | April | May | |
| X: | 1 | 2 | 3 | 4 | 5 |
| Y: | 25 Cars | 28 Cars | 30 Cars | 32 Cars | 40 Cars |
| Solution | ||
|---|---|---|
| ?x | = | 15 |
| ?x 2 | = | 55 |
| ?y | = | 155 |
| ?y 2 | = | 4,933 |
| ?xy | = | 499 |
| n | = | 5 |
Curve fitting is a method for mathematically describing the relationship between two variables. The equation of...
