Shared Flashcard Set

Details

S14.1 - Testing Significance, Least-Squares Regression Model
Testing the Significance of the Least-Squares Regression Model
26
Mathematics
Graduate
11/21/2013

Additional Mathematics Flashcards

 


 

Cards

Term
1. Requirement for inference on the least-squares regression model
Definition
The population is linear on x
mew (ylx) = B1x + B0
Term
2. Requirement for inference on the least-squares regression model
Definition
The response variables are normally distributed with mean mew (ylx) = B1x + B0
Term
Least-squares regression model
Definition
yi = B1xi + B0 + ei
Term
Standard error of the estimate
Definition
The unbiased estimator of sigma
Term
Standard error of the estimate, se =
Definition
[(summation(yi-yihat)^2)/n-2]^0.5
Term
The denominator of the standard error of the estimate is
Definition
n-2 because 2 parameters (betas) cause it to lose 2 degrees of freedom
Term
Because the response variable needs to be normally distributed for a least-squares regression, this means that
Definition
ei, the residuals, must also be normally distributed
Term
If there is no linear relationship between the response and explanatory variables, the slope of the true regression line will be
Definition
0
Term
Two-tailed test, least-squares regression
Definition
Testing whether a linear relation exists between two variables without regard to the slope
Term
Left-tailed test, least-squares regression
Definition
Testing whether the slope of the true regression line is negative
Term
Right-tailed test, least-squares regression
Definition
Testing whether the slope of the true regression line is positive
Term
Least-squares regression model, t-distribution =
Definition
b1 - B1/sb1
Term
Least-squares regression model, sb1 =
Definition
[se/(summation(xi-xbar)^2)^0.5]
Term
2 requirements for hypothesis test regarding the slope coefficient, B1
Definition
1. The sample is obtained using random sampling
2. The residuals are normally distributed with constant error variance
Term
Two-tailed test, slope coefficient
H0: p =
H1: p =
Definition
H0: B1 = 0
H1 B1 not=to 0
Term
Left-tailed test, slope coefficient
H0: p =
H1: p =
Definition
H0: B1 = 0
H1 B1 < 0
Term
Right-tailed test, slope coefficient
H0: p =
H1: p =
Definition
H0: B1 = 0
H1 B1 > 0
Term
Two-tailed test, classical approach
If t0 > t(alpha/2)
Definition
Reject the null
Term
Left-tailed test, classical approach
If t0 < -t(alpha)
Definition
Reject the null
Term
Right-tailed test, classical approach
If t0 > t(alpha)
Definition
Reject the null
Term
P-value approach
If p-value < alpha
Definition
Reject the null
Term
The slope coefficient hypothesis testing is considered
Definition
Robust
Term
The confidence interval for the slope of the least-squares regression line is of the form
Definition
Point estimate +/- margin of error
Term
Lower bound CI, slope of the regression line =
Definition
b1 - t(alpha/2)[se/(summation(xi-xbar)^2)]
Term
Upper bound CI, slope of the regression line =
Definition
b1 + t(alpha/2)[se/(summation(xi-xbar)^2)]
Term
Bivariate normal distribution/Jointly normal distributed
Definition
When the y's at any given x and the x's at any given y are normally distributed
Supporting users have an ad free experience!