Lernmaterialien für ökonometrie an der Humboldt-Universität zu Berlin

• same as usual
• perform hypothesis test on coefficients of dummy variables 

• complete set off all causal relationships in a graph with arrows
• help explain OMV, reverse causality and conditioning on colliders
• arrows represent direct effect

• errors occur in poulation and are never actually observed
• residuals occur in our model and are computed from the data

• if the regression model is to have j intercepts for j different groups w need to include j-1 dummy vars.
• otherwise violation of MLR.3
• base group: overall intercept of model
• dummy variable coefficients: differences in intercepts between this group and base group

• functional forms as discussed in MLR work as before
• dummy variables work as before

• consists of time series for each cross sectional data in the set

Development of statistical methods for estimating economic relationships.

• if y is multiplied by c then intercept and slope estimate are multiplied by c

• if x is multiplied by c then slope is divided by c

  
  

• if y+m intercept changes by m

• if m is added to x the intercept changes by m*β1, slope does not change

All other factors fixed

• consists of random sample taken at a given time
• order of data irrelevant