Ordinary Least Squares Flashcards
time-series data
data that follows one person, country, firm, or other entity across different periods of time.
cross-section data
data that contains information for many people, countries, firms, or entities for the same period of time.
residual
the observed error term.
calculated by subtracting the actual Y-value from the predicted Y-value.
ordinary least squares (OLS)
method for estimating a regression line in which values are found for the coefficient estimates that minimize the sum of “e-hat squared”
multivariate regression model
a regression model with more than one independent variable.
extrapolation
projecting beyond the scope of the project.
degrees of freedom
the degrees of freedom provide a measure for the amount of information that is available to estimate a regression model.
they are equal to the sample size minus the number of coefficients that need to be estimated in the model (n - k - 1)
one degree of freedom is subtracted off for the intercept hence the “-1”
The Central Limit Theorem
the mean of independently and identically distributed random values will be normally distributed as the sample size becomes larger.
selection bias
occurs when a particular group of observations is underrepresented in a sample.
sample
consists of members of a group that are chosen to represent the whole group.
population
includes all members of the group.
significance level
the significance level is the probability of a Type I error that you are willing to accept when you test your hypothesis.
critical value
in general, a critical value is the value of a test statistic that marks the beginning of the rejection region.
t-test
a t-test is the most common test in econometrics. it helps assess the chances of a slope’s true value being zero.
decision rule
a decision rule tells you how to use a test statistic to decide whether to reject the null hypothesis.
