week02-1

week02-1 - TA session 2 Econ 103 winter 2010 Wed Jan 13...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
TA session 2 Econ. 103, winter 2010 Wed., Jan. 13, 2010, 10:00 a.m. and 1:00 p.m. in PP2400E. 1 Thinking statistically: when models go wrong indep. variables dep. variable cause efect (model) In econometrics (and in statistics generally) the modeler hopes to link a set of causes expressed quantitatively ( e.g. age, education, experience) with an eFect expressed quanti- tatively ( e.g. the wage). I have illustrated the idea above. The line connecting cause with eFect is the formal statistical model . In statistics it is always important to understand the hidden assumptions one makes in connecting cause with eFect in any model. Example: In this class Prof. Casanova will teach you the linear regression model . A key assumption of this model, as seen in its name, is that the independent variables have a linear eFect on the dependent variable. 2 Problems with model causality If ever you are asked to critique a statistical model and you don’t know what to say, just think to yourself “correlation is not causation”
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 2

week02-1 - TA session 2 Econ 103 winter 2010 Wed Jan 13...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online