Question

3. Knowing that a university has the following units/campuses: A, B , the medical school
in another City. You are interested to know on average how many hours per week the university
students spend doing homework. You go to A campus and randomly survey students walking
to classes for one day. Then,this is a random sample representing the entire
university students population.
True False
4. The Law of Large Number(LLN)is related with the concept of convergence in probability, while The
Central Limit Theorem(CLT)is related with convergence in distribution.
True False
5.You have a cross-sectional dataset with an independent variable X and a dependent variable Y.You
find a positive correlation between X and Y.Then you can conclude that X causes Y.
True False
6. In a cross-sectional dataset the order of the observations is arbitrary,while in a time series dataset the
order is important because it is likely that we have correlated observations.
True False
7. Consider the following simple linear regression model:y=Bo+B1t+u.The essential assumption to
derive the estimators of Bo and B1 through the Method of Moments is E(u|X)=0.
True False

8. Consider the following simple linear regression model: y=B0+B1x+u. when we derive the estimators for B0 and B1 we get 2 foc

True False

Answer 1

3. False. - That random sample collect the informations associated with only A campus students. The informations of the students from B campus are missing. Therefore, the sample does not represent the entire university population.  

4. True - The law of large number is a result in probability theory also known as Bernoulli's theorem. Whereas, the central limit theorem is associated with the probability distribution of sample mean.  

5. False - If X is the independent variable and Y is the dependent variable then any positive correlation between X and Y conclude that Y causes X. Here Y is the causal variable.

6. True - A cross section data set consists of the informations at a point of time. So these data set may be arbitrary. However, a time series data set consists of information on a variable over time. Therefore, the information on a variable over time are correlated so their order is important.    

7. False - The method of moments is based on a simple principle which states that one should estimate a moment of the population distribution by the corresponding moment of the sample. The assumption that E(u) = 0 is associated with the specification of the model.  

8. False - Consider the following simple linear regression model: y=B0+B1x+u. when we derive the estimators for B0 and B1 we get two normal equations.