12: Multiple Regression and Issues in Regression Ana

luqian55

UID: 118990
帖子: 774
主题: 1
注册时间: 2008-12-31
最后登录: 2012-9-5

17^#

发表于 2010-5-30 06:07 | 只看该作者

thanks

maxsimax

UID: 107737
帖子: 1401
主题: 0
注册时间: 2008-9-6
最后登录: 2016-9-21

16^#

发表于 2010-4-14 16:32 | 只看该作者

thanks

UID: 137525
帖子: 5724
主题: 885
注册时间: 2009-7-1
最后登录: 2011-3-29

15^#

发表于 2010-4-8 15:04 | 只看该作者

An analyst is trying to determine whether fund return performance is persistent. The analyst divides funds into three groups based on whether their return performance was in the top third (group 1), middle third (group 2), or bottom third (group 3) during the previous year. The manager then creates the following equation: R = a + b₁D₁ + b₂D₂ + b₃D₃ + ε, where R is return premium on the fund (the return minus the return on the S& 500 benchmark) and D_i is equal to 1 if the fund is in group i. Assuming no other information, this equation will suffer from:

A)	collinearity.

B)	heteroskedasticity.

C)	serial correlation.

There are too many dummy variables specified, so the equation will suffer from collinearity.

UID: 137525
帖子: 5724
主题: 885
注册时间: 2009-7-1
最后登录: 2011-3-29

14^#

发表于 2010-4-8 15:04 | 只看该作者

Suppose the analyst wants to add a dummy variable for whether a person has an undergraduate college degree and a graduate degree. What is the correct representation if a person has both degrees?

Undergraduate Degree Graduate Degree

Dummy Variable Dummy Variable

0 0

1 1

0 1

UID: 137525
帖子: 5724
主题: 885
注册时间: 2009-7-1
最后登录: 2011-3-29

13^#

发表于 2010-4-8 15:04 | 只看该作者

Suppose the analyst wants to add a dummy variable for whether a person has an undergraduate college degree and a graduate degree. What is the correct representation if a person has both degrees?

Undergraduate Degree Graduate Degree

Dummy Variable Dummy Variable

0 0

1 1

0 1

Assigning a zero to both categories is appropriate for someone with neither degree. Assigning one to the undergraduate category and zero to the graduate category is appropriate for someone with only an undergraduate degree. Assigning zero to the undergraduate category and one to the graduate category is appropriate for someone with only a graduate degree. Assigning a one to both categories is correct since it reflects the possession of both degrees.

UID: 137525
帖子: 5724
主题: 885
注册时间: 2009-7-1
最后登录: 2011-3-29

12^#

发表于 2010-4-8 15:04 | 只看该作者

The amount of the State of Florida’s total revenue that is allocated to the education budget is believed to be dependent upon the total revenue for the year and the political party that controls the state legislature. Which of the following regression models is most appropriate for capturing the effect of the political party on the education budget? Assume Y_t is the amount of the education budget for Florida in year t, X is Florida’s total revenue in year t, and D_t = {1 if the legislature has a Democratic majority in year t, 0 otherwise}.

A)	Y_t = b₁D_t + b₂X_t + e_t._{> >}

B)	Y_t = b₀ + b₁D_t + b₂X_t + e_t.

C)	Y_t = b₀ + b₁D_t + e_t._{> >}

UID: 137525
帖子: 5724
主题: 885
注册时间: 2009-7-1
最后登录: 2011-3-29

11^#

发表于 2010-4-8 15:04 | 只看该作者

A)	Y_t = b₁D_t + b₂X_t + e_t.

B)	Y_t = b₀ + b₁D_t + b₂X_t + e_t.

C)	Y_t = b₀ + b₁D_t + e_t.

In this application, b₀, b₁, and b₂ are estimated by regressing Y_t against a constant, D_t, and X_t. The estimated relationships for the two parties are:

Non-Democrats:	? = b₀ + b₂X_t
Democrats:	? = (b₀ + b₁) + b₂X_t

UID: 137525
帖子: 5724
主题: 885
注册时间: 2009-7-1
最后登录: 2011-3-29

10^#

发表于 2010-4-8 15:03 | 只看该作者

Consider the following model of earnings (EPS) regressed against dummy variables for the quarters:

EPS_t = α + β₁Q_1t + β₂Q_2t + β₃Q_3t

where:
EPS_t is a quarterly observation of earnings per share
Q_1t takes on a value of 1 if period t is the second quarter, 0 otherwise
Q_2t takes on a value of 1 if period t is the third quarter, 0 otherwise
Q_3t takes on a value of 1 if period t is the fourth quarter, 0 otherwise

Which of the following statements regarding this model is most accurate? The:

A)	EPS for the first quarter is represented by the residual.

B)	coefficient on each dummy tells us about the difference in earnings per share between the respective quarter and the one left out (first quarter in this case).

C)	significance of the coefficients cannot be interpreted in the case of dummy variables.