TN - Endogeneity (2) - Endogeneity The technical definition of endogeneity is that one of the independent variables(e.g Xs is correlated with the error

TN - Endogeneity (2) - Endogeneity The technical definition...

This preview shows page 1 - 3 out of 6 pages.

Endogeneity The technical definition of endogeneity is that one of the independent variables (e.g., X’s) is correlated with the error term, i.e., cov ( X j , ε ) 0 for some j. In a practical sense, the presence of endogeneity means that the stand regression estimates are biased in inconsistent (i.e., you can not use them). So, in this module we discuss the ways to test for and correct for this problem. Three common reasons for endogeneity are: 1) Missing variables 2) Time Frame 3) Simultaneity (e.g., using one output to predict another) Missing Variables and Proxy Variables One key assumption of regression is commonly called a “true model” assumption, e.g., the model under consideration is an accurate representation of the underlying process generating the data, e.g., all of the key predictor variables are included in the model. However, the true assumption is that any variables lef out of the regression are uncorrelated with the predictors included in the model. For instance, consider the following equation (and assume that this is the true model): ln ( Wage ) = β 0 + β 1 Education + β 2 Ability + ε It is also reasonable to assume that Education is correlated with Ability. So, if we run the regression without Ability: ln ( Wage ) = β 0 + β 1 Education + ε' then, ability is included in the new error term ’, and we have an endogeneity problem, i.e., the estimates will be biased because Education is correlated with the error term (which includes all omitted variables). If we do not have a true measure of ability, we can used what is called a Proxy variable to fix this problem. A proxy variable is one that is strongly correlated with Ability, e.g., IQ. So, we could run the following regression to fix the endogeneity problem: ln ( Wage ) = β 0 + β 1 Education + β 2 IQ + ε where IQ is the proxy for Ability. Seasonality and time trend (e.g., a company sales are growing over time) are other examples of situations where omitted variables can cause endogeneity problems. For instance, consider the following graph of a company’s total transactions and catalogue transactions over time. Clearly, there are large spikes of sales that occur at the end of year indicating that the sales are very seasonal. If marketing activities are also seasonal (i.e., more intense marketing during the spiked periods), omitting seasonality from the model will bias the marketing effects.
Image of page 1

Subscribe to view the full document.

0 50 100 150 200 250 300 350 total_trans catalogue Week Transactions Time Frame A second type of endogeneity present in many datasets is the time-frame of the data versus the time- frame of managerial decision making. What this means is whether or not a manager can adjust his/her decision variables within the time-frame of the data. For instance, grocery stores in the U.S. change prices on a weekly-basis (due to them sending out fliers, etc.). So, if the data is at the weekly-level, then price can be treated as an exogenous (i.e., not endogenous) variable. However, if the data is at the monthly-level, then the manager can adjust price in response to local demand. This means that price is endogenous (i.e., an output of the system).
Image of page 2
Image of page 3

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern

Ask Expert Tutors You can ask 0 bonus questions You can ask 0 questions (0 expire soon) You can ask 0 questions (will expire )
Answers in as fast as 15 minutes