ma15 - A = the intercept of the regression line with the...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Quantitative Forecasting Methods (Trend Projections and Regression) If we have a number of historical data points, we can fit a trend line to those points to make predictions about the future. (Note that the regression technique that we will discuss in this section only applies if the trend is linear.) Linear trends can be found by using the least squares technique. This technique fits a “best fit” line to a series of points by minimizing the sum of the squared errors (deviation between the points and the line). These deviations are assumed to be due to random causes. The process of running a regression using the least squares method is made simple using the following formulas: Y^ = a + bx b = the sum of (xy) – n (x*) (y*) / the sum of (x-squared) – n (x*squared) a = y* - b(x*) Where… Y^ = the computed value of the dependent variable X = the independent variable
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: A = the intercept of the regression line with the y-axis B = the slope of the regression line N = the number of periods X* = the average of the x’s Y* = the average of the y’s It is crucial to remember that there must be a linear relationship before a useful trend line can be estimated, and we have to assume that deviations around this linear trend are random. To ensure that we have a linear relationship, it is always best to plot the data first. Also, we do not want to use the resulting trend line to make predictions too far out in the future, because the linear relationship may change over time. For example, you certainly would not want to use a linear relationship showing sales of cathode-ray television sets from 1985 to 1995 in order to predict the demand for cathode-ray television sets in 2011!...
View Full Document

This note was uploaded on 02/03/2011 for the course MAN 4504 taught by Professor Benson during the Spring '08 term at University of Florida.

Page1 / 2

ma15 - A = the intercept of the regression line with the...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online