Session 7 - Chapter 5 Flashcards
How do we summarize the relationship between these two variables ?
if the price increase, the sales decrease.
We can represent it with a decression line
What is a dichotomous treatment ?
Dichotomous treatment Two treatment statuses—treated and untreated
What is a regression analysis ?
A regression analysis = The process of using a function to describe the relationship among variables
A quoi est égal m ?
m = Y1-Y0/X1-X0
Quelle est la procédure ?
0) On sépare les données en deux, on dessine les données sur un graphe.
1) On imagine une fonction Y = f(X) = b + m (X)
2) On remplace x par 0 ou par 1.
3) On calcule b, la moyenne des profits des semaines non-traitées.
4) On calcule m, m = Y1-Y0/X1-X0
Donne une formule simplifiée
Outcome = Untreated mean + (Treated mean - Untreated mean) * Treatment
What is the residual “e” ?
e = Y - f(x)
The residual is the difference between the observed outcome and the corresponding point on the regression line for a given observation.
What is our objective ?
The objective is to minimize these residuals.
What are the 2 conditions the “ideal” point- line satisfy ?
1) It should not generally overshoot or undershoot the data
2) Its tendency to over or undershoot the data across specific price levels should not depend on the price level
How many regression parameters will be estimated in a simple linear regression model?
You estimate the equation of the line (b + m (x)), so you have x, you have to estimate 2 parameters b and m. More the two level of treatment, is unknown.
What is the calcul of m with sCov and sVar ?
m = sCov (X,Y)/sVar
What is OLS ?
Ordinary Least Squares
It is the process of solving for the slope and intercept that minimize the sum of the squared residuals.
What is LAD ?
Least absolute deviations (treated by the absolute value)
Use the sum of the absolute value of the residuals as the objective function and solve for the slope and intercept that minimize it
What is an objective function ?
An objective function is best described as a function ultimately wished to be maximized or minimized.
What is a multiple regression ?
Multiple regression = Solving for a function that best describes the data the implies the use of OLS (or equivalently, the sample moment equations)
