Downloads
cwp411515.pdf
PDF | 603.45 KB
This paper considers the finite sample distribution of the 2SLS estimator and derives bounds on its exact bias in the presence of weak and/or many instruments. We then contrast the behavior of the exact bias expressions and the asymptotic expansions currently popular in the literature, including a consideration of the no-moment problem exhibited by many Nagar-type estimators. After deriving a finite sample unbiased k-class estimator, we introduce a double k-class estimator based on Nagar (1962) that dominates k-class estimators (including 2SLS), especially in the cases of weak and/or many instruments. We demonstrate these properties in Monte Carlo simulations showing that our preferred estimators outperforms Fuller (1977) estimators in terms of mean bias and MSE.
Authors
Working Paper details
- DOI
- 10.1920/wp.cem.2015.4115
- Publisher
- Institute for Fiscal Studies
Suggested citation
M, Harding and J, Hausman and C, Palmer. (2015). Finite sample bias corrected IV estimation for weak and many instruments. London: Institute for Fiscal Studies. Available at: https://ifs.org.uk/publications/finite-sample-bias-corrected-iv-estimation-weak-and-many-instruments (accessed: 28 March 2024).
More from IFS
Understand this issue
Where next for the state pension?
13 December 2023
Social mobility and wealth
12 December 2023
Autumn Statement 2023: IFS analysis
23 November 2023
Policy analysis
Recent trends in public sector pay
26 March 2024
Living standards since the last election
21 March 2024
Major challenges for education in Wales
21 March 2024
Academic research
Social skills and the individual wage growth of less educated workers
27 March 2024
House price rises and borrowing to invest
27 March 2024
Household responses to trade shocks
26 March 2024