|
III. Regression and Time Series Analysis
A. Correlation and Linear
Regression
1. Correlation
Measure the
strength of the association (relationship)
between two variables
a.
Characterization
.
.
.
.
.
.
.
.
.
.
b. Calculation
.
.
.
.
.
.
.
c. Hypothesis test
.
.
.
.
.
.
2. Linear regression
.
.
.
a. Simple linear
regression model
.
.
.
.
.
.
.
.
.
b. Methodology
- Least squares
method - minimize sum of squared
differences of the actual values from the
estimated values
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
c. Interpretation
.
.
.
.
.
.
.
.
d. Measures of
variation
(1) Sum of squares
.
.
.
.
.
.
.
.
.
.
(a) Total
sum of squares (SST)
.
.
(b)
Regression sum of squares (SSR)
.
.
(c) Error
sum of squares (SSE)
.
.
(2)
Coefficient of determination
- Proportion
of the variation in variable Y explained
by the linear regression model involving
variable X
.
.
.
.
.
.
.
(3) Standard
error
- Standard
deviation of actual values of Y around
the predicted values of Y
.
.
.
e. Prediction
.
.
.
.
|
