This preview shows page 1. Sign up to view the full content.
Regression Terminology: Deterministic modes

Regression formulas that predict an exact
value.
Probabilistic modes
Regression formulas that take into account random error, and the
fact that other variables account for some of the variation in the values of y. What we use in
statistics.
Residual
Error of the regression line at a particular point. Difference between y (real
y) and ŷ (predicted y).
Heteroscedasticity
Error variance that is no constant.
Homoscedasticity

Constant error variance.
Standard error of the estimate (s
e
)
A measure of how much error is in
the regression model. Need to have a lot of
contend knowledge
of what is being estimated to
fully use the
s
e
.
Sum of Squares Error (SSE = ∑(y  ŷ)
2
)

Some of the
types of Regression Analysis
 we are trying to predict
y
and
x
.
Simple Regression
Analysis
: predicting a continuous
y
variable from a continuous
x
variable.
Logistic Regression
Analysis
: predicting a dichotomous
y
variable from one or more continuous
x
variables.
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 01/26/2012 for the course QMST 2333 taught by Professor Mendez during the Spring '08 term at Texas State.
 Spring '08
 Mendez

Click to edit the document details