Metrics Mondays

(Note: The most recent posts will be added to the top of the list, and posts are generally listed from most recent to oldest.)

  1. Generated Regressors, or Why Regressing on \hat{X} Can Be a Problem
  2. Good Things Come to Those Who Weight–Part I
  3. Regression and Causality for Dummies
  4. Achieving Statistical Significance with Covariates
  5. Dealing with Imperfect Instruments III
  6. We Wrote a Paper About Lagged Explanatory Variables. Here’s What Happened Next.
  7. Interactions as IVs and Spurious Findings
  8. One IV for Two Endogenous Variable, and Testing for Mechanisms
  9. How Should Econometrics Be Taught?
  10. Combining Bits and Pieces of Likelihood to Study Behavior
  11. Fixed Effects, Random Effects, and (Lack of) External Validity
  12. Dealing with Duration Data
  13. Heteroskedasticity and Its Content
  14. Dealing with Imperfect Instruments II
  15. Dealing with Imperfect Instruments I
  16. Testing for Mechanisms (and Possibly Ruling Out All Other Mechanisms)
  17. How to Systematically Think about Selection
  18. Simpson’s Paradox, or Why “Determinants of …” Papers are Problematic
  19. Lagged Explanatory Variables and the Estimation of Causal Effects
  20. Estimating Nonlinear Relationships
  21. Nothing Compares 2 U
  22. Robustness Check or Data Mining?
  23. What to Do with Repeated Cross Sections?
  24. Interpreting Coefficients II
  25. “Are Those Two Distributions Alike?” Redux
  26. Interpreting Coefficients I
  27. Are Those Two Distributions Alike?
  28. What to do When You Have the Whole Distribution Instead of a Sample?
  29. Statistical vs. Economic Significance
  30. Type III Errors
  31. The Tobit Temptation
  32. There Is More than One Source of Endogeneity
  33. Why You Should Show a Regression of Y on Z
  34. Fads and Fashions in Econometrics
  35. Multicollinearity
  36. Friends *Do* Let Friends Do IV
  37. Regressions as Ecosystems
  38. When Is Heteroskedasticity (Not) a Problem?
  39. Hypothesis Testing in Theory and in Practice
  40. Statistical Literacy
  41. Data Cleaning
  42. Outliers
  43. Proxy Variables
  44. What to Do with Missing Data
  45. What to Do with Endogenous Control Variables
  46. Control Variables: More Isn’t Necessarily Better
  47. You Can’t Test for Exogeneity: Uninformative Hausman Tests
  48. “Do Both”
  49. You Keep Using that Instrumental Variable; I Do Not Think It Does What You Think It Does
  50. PSA: p-Values Are Thresholds, Not Approximations
  51. The Use and Misuse of R-Square
  52. Big Dumb Data?
  53. Rookie Mistakes in Empirical Analysis
  54. Goodness of Fit in Binary Choice Models
  55. A Nifty Fix for When Your Treatment Variable Is Measured with Error
  56. A Rant on Estimation with Binary Dependent Variables
  57. Love It or Logit, or: Man, People *Really* Care about Binary Dependent Variables
  58. In Defense of the Cookbook Approach to Econometrics
  59. More on the Cookbook Approach to Econometrics
  60. Econometrics Teaching Needs an Overhaul
  61. Hipstermetrics
  62. On the (Mis)Use of Regression Analysis: Country Music and Suicide
  63. Methodological Convergence in the Social Sciences