Class Notes Flashcards
Three Major Uses of Econometrics
- ) Describing Economic Reality
- ) Testing hypotheses about economic theory
- ) Forecasting future economic activity
Econometrics is…
the quantitative measurement and analysis of actual economic and business phenomenon
Regression Analysis is…
statistical technique that attempts to “explain” movements in one variable, the dependent variable, as a function of movements in a set of other variables, called the independent (or explanatory) variables, though the quantification of a single equation.
What are three Alternative Econometric Approaches
- ) Specifying the models or relationships to be studied
- ) Collecting the data to quantify the models
- ) Quantifying the models w/ data
Stochastic Error Term
is a term that is added to a regression equation to introduce all the variation in Y that cannot be explained by the included x
Stochastic Error Term is a symbol of …
the econometrician’s ignorance or inability to model all the movements of the dependent variable
Stochastic error term must be present in a regression equation because…
there are at least four other sources of variation in Y variation in the included X’s
Key Assumptions about the error term:
- ) The variation of e does not change when x changes
- ) The problem distribution of e is normal
- ) Estimated sum of e is 0
- ) The value of e is independent of e for any two observations I and j
Test whether the slope coefficient is significantly different from zero
Step 1: estimate variance, to do this we need to know how widely the error is distributed.
Step 2: We use the available data to estimate
a.) Compute a confidence interval around slope coefficients where Null = B1 = 0
b.) compute the T - statistic for the difference btw B1 and 0
OLS
- Ordinary Least Squares, also known as minimizing squared errors
P value and T statistics are unitless so
always factor in the difference btw substantive vs. statistic difference
TSS =
RSS + ESS
Why should we include more than one variable in our regression?
- ) To improve the precision of our predictions of Y
2. ) To avoid biased estimate of the co-efficients (“omitted variable bias”)
Multicolinearity
is when your Rk2 is big and results in large standard errors (multivariate)
The standard error reported in STATA is…
s^2 = RSS/(N-(k+1))