ACCAspace_sitemap
PPclass_sitemap
sitemap_google
sitemap_baidu
CFA Forums
返回列表 发帖
 

6、Linear regression is based on a number of assumptions. Which of the following is least likely an assumption of linear regression?

A) Values of the independent variable are not correlated with the error term.

B) A linear relationship exists between the dependent and independent variables.

C) There is at least some correlation between the error terms from one observation to the next.

D) The variance of the error terms each period remains the same.

TOP

 

The correct answer is

When correlation exists, autocorrelation is present. As a result, residual terms are not normally distributed. This is inconsistent with linear regression.


TOP

 

7、Which of the following is least likely an assumption of a simple regression?

A) The variance of the error term is one.

B) The error term is normally distributed. 

C) The expected value of the error term is zero. 

D) There is a linear relationship between dependent and independent variables.

TOP

 

The correct answer is A

There is no requirement that the variance of the error term should be equal to one.


TOP

 

AIM 2: Define and distinguish homoskedasticity and heteroskedasticity.

Which expression best represents the condition homoskedasticity? (In the expressions assume σ2 > 0)

A) V(εi|Xi) = σ2. 

B) E(εi|Xi) = σ2.

C) corr(Xi, εi) = 0.

D) corr(εi, εi + j) = 0. 

TOP

 

The correct answer is A

Homoskedasticity means the variance of εi is constant and unrelated to the value of the independent variable.


TOP

 

AIM 3: Define, calculate and interpret the standard errors of the coefficients in an OLS model.

In a two-variable regression of the dependent variable Y on the independent variable X, the standard error of the slope coefficient is:

 

A )

  1.gif

 

B)

  2.gif

 

C)

  3.gif


 

D)

4.gif


[此贴子已经被作者于2009-6-26 11:05:41编辑过]

TOP

 

The correct answer is C

The correct formulas for the standard errors uses X and not Y. The other formula among the choices using X is the standard error of the intercept.

 

TOP

 

AIM 5: Explain the application of the Gauss-Markov and Central Limit Theorem in OLS estimates.

1、The Gauss-Markov theorem says that if the classical linear regression model assumptions are true, then the OLS estimators have all of the following properties except the:

A) OLS estimated coefficients are based upon linear functions.

B) OLS estimated coefficients are unbiased, which means E(b0) = B0 and E(b1) = B1.

C) OLS estimate of the variance of the errors is unbiased, i.e., E()= σ2.

D) OLS estimated coefficients have the minimum absolute error when compared to other methods of estimating the coefficients, i.e., they are the most precise.

TOP

 

The correct answer is D

They have the minimum variance, which is not the same as the minimum absolute error.


TOP

返回列表