Dynamic Models for Dynamic Theories: The Ins and Outs of Lagged Dependent Variables

By Luke Keele, N. Kelly. Political Analysis, 2006

922 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Political Science and International Relations

View Paper: https://www.semanticscholar.org/paper/ea0e69b5eeefdd8a022704c5e2a543bd1e9ce646

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

1. Haiyan Song
2. G. Crouch
3. Bo Peng
4. Stephen F. Witt
5. S. Heichel
6. Gang Li
7. Christopher Wlezien
8. M. Debus

In Political Science and International Relations

1. Christoph Knill
2. S. Heichel
3. Christopher Wlezien
4. M. Debus
5. Nathaniel Becky
6. Moira Burke
7. Dursun Peksen
8. A. Drury

Citations in the corpus

• How Well Do Social Ratings Actually Measure Corporate Social Responsibility? (2008, positive): “Including a lagged dependent variable is a common way to control for such autocorrelation (Keele and Kelly, 2006).”
• Density and Dispersion: The Co-Development of Land Use and Rail in London (2007, result): “This finding was supported by Keele and Kelly (2006) who say 'if the process was dynamic, OLS with a [lagged dependent variable] provided estimates that were superior to the other models or estimators even in the presence of minor residual autocorrelation.”
• The power of people: social capital’s role in recovery from the 1995 Kobe earthquake (2011, positive): “Known as the ‘‘workhorse’’ of time-series models ( Keele and Kelly 2005 ), this model can adopt values of i (the unit) from 1 to 9 (for the 9 wards within Kobe) and values of t (time, measured in years) from 2 to 18 (as the dataset contains 18 years of measurements).”
• Social capital on facebook: differentiating uses and users (2011, background): “Lagged independent variables (previous site activity) are not included because they are highly collinear, and thus would produce biased estimates (see [21]).”
• Modeling Dynamics in Time-SeriesCross-Section Political Economy Data (2011, methodology): “In our prior work (Beck & Katz 1996) we argued that a lagged dependent variable (LDV) specification is often adequate; because that has sparked some discussion (Achen 2000, Keele & Kelly 2006), we spend some time on this issue.”
• Growing closer on facebook: changes in tie strength through social network site use (2014, background): “Lagged independent variables (communication the previous month) are not included because they are highly collinear, and thus would produce biased estimates [20].”