a) Econometrics is the development of statistical method for estimating economic relationships, testing economic theories, and evaluating and implementing government and business policy. It is the branch of economics that aims to give empirical content to economic relations, allowing theories to be tested and used.
Estimate the parameters in a model and to test hypotheses about these parameters.
Using these parameters to determine the validity of an economic theory and the effects of certain policies.
Examine how well an economic theory can explain historical economic data (particularly the important facts)
Predict the future evolution of the economy
c) 1. Statement of theory or hypothesis.
2. Specification of the mathematical model of the theory
3. Specification of the econometric model of the theory
4. Obtainment of data
5. Estimation of the parameters of the econometric model
6. Hypothesis testing
7. Forecasting or predicting
8. Using the model for control or policy purposes.
d) Mathematical Economics reveals economic activity by using theoretical relationships between various factors and uses mathematical equations to describe certainty. Experimental data is used in Mathematical Economics.
On the other hand, Econometrics focuses on the problems inherent in collecting and analyzing nonexperimental data. It reveals the quantitative relationships between various factors and uses mathematical equations to describe randomness.
Economics and social science data are mostly found in time-series form (covering more than one time period) or in cross-sectional data (in one time period) or it may be collected from surveys.
For random sampling, it is sometimes not appropriate as an assumption for analyzing cross-sectional data because of sample selection problem in which biased data might be collected. Another violation of random sampling occurs when the sample unit is large relative to the population
For time-series, economic observations can rarely be assumed to be independent of time. Most economic and other time series have a strong correlation to their recent histories. Data frequency and in time series is an important factor to take into consideration in the treatment of seasonal time series.
a) Cross sectional data as many subjects are observed (U.S states) at the same point in time (March 2006).
b) Time series as it consists of the observation of one variable (inflation in India) over time.
c) It is a time-series as the data set consists of two variables (minimum wage changes and teenage employment) over time (1980-2004).
d) Assuming that the students are chosen randomly and since all variables but that the control (computer usage) are held constant, this dataset would be categorized as a randomized controlled experiment.
e) It is a cross-sectional data set consistent of a sample of individuals (members of Congress) at a given point in time.
• Number of police officers/patrols existent
• Number of Years of schooling
• Criminal Punishment in Years of jail time
• Criminal Punishment in terms of value of fines to be paid
• Family conditions: Divorce rate
• Unemployment Rate
• Youth concentration of community
• Crime reporting practice of community
• Poverty level
• I would design an experiment in which 100 applications for housing are sent for processing. 50 of these applications will be representative of French resident and the other 50 will be constituted of several racial minorities. In order to hold all other factors constant, the applications will all be the same except for any information or proxies indicative of race (name, race, nationality).
Half of the French applications will be for neighborhoods dominated by the French, the other half for neighborhoods dominated by minorities. Likewise, half of the applications of minorities will be sent for French neighborhoods, and the other half for neighborhoods with minorities.
• As proposed in the experiment suggested above, the dominant race in the target neighborhood might have an effect in the success rate of the candidate to rent an apartment as a method of segregation. Racism toward some ethnic minorities is another factor to be taken into consideration.
a) Most likely, one would find a positive correlation between the returns to education for individuals who choose to go to college, ceteris paribus. Therefore, A will receive a higher average wage than B.
b) Correlation does not imply causation. Other variables come into work in the determination of wage. For example, people who attend college are possibly more likely to be me harder working of more gifted than those who don’t. However, if in the case of control, one can find causal effect between college attendance and return as one prompts the other directly.
c) Finding twins for example, one of which attended college and another who did not attend. Examine wage differentials between the two. Assume here that twins are identical except for the fact that one attended college and the other didn’t.
Another example would be focusing on the same subject. Examine earnings of this individual before attending college and after (if possible). Then examine wage differential. Assumption here is that the only thing that has changed is the acquiring of a college data and that there is no change in personal abilities, productivity, etc.