Econometric tools and techniques

The warchest

Murtaza Haider

Outline

  • Table 0

  • Table 1

  • Visualizing data

    • A good plot is worth a pub in A journal.

      • Data Mapper
    • Statistical tools

      • Tests

      • Models

      • Predictions

Table 0 (Often in Appendix)

  • Variable names and description

  • Saves editors and reviewers’ time by advising them of what variables represent.

Table 1

The first table in the manuscript

  • Empirics start with Table 1.

  • After the abstract, Table 1 is the most critical part of the paper.

  • The first and most important table in the manuscript

  • Sets the stage for the evidence that follows

  • Should present sufficient evidence to necessitate the need for further inquiry

  • Must present only relevant variables and appropriate summary statistics

    • Means and Standard Deviations for continuous variables

    • Frequencies and percentages for categorical variables

  • Show tests if necessary

Options for tables

  • htsummary (my favourite)

  • dtable (new in Stata 18)

  • See examples

HTSummary

https://journals.sagepub.com/doi/pdf/10.1177/1536867X1201200409

15-minute city

  • Old wine in new bottles. A place where most destinations are reachable in 15 minutes by sustainable modes and public transit.

  • The Challenge: Demographics and location

  • Downtown by another name

Downtown by another name

15-minute neighbourhood demographics

dtable

webuse nhanes2
dtable age weight bpsystol i.sex i.race, by(diabetes, tests)

Charts and graphs

Bars as Columns

Flipping the Columns

Comparison (bar charts)

Comparison 2 (bar chart)

Dramatic

The story in one chart

Condofication of a metropolis

Transitions - 1

Transitions - 2

Maps

Mapping migrations

Statistical Tools

  • Depending upon the type of dependent variable

    • Continuous

    • Categorical

  • Depending upon the structure of data

    • Cross-section

    • Panel

    • Time Series

    • Panel Time Series

Cross-sectional data

  • Ordinary Least Squares Regression (OLS)

  • Generalized Least Squares

    • Dichotomous DV (Logit / Probit)

    • Ordinal DV (Ordinal)

    • Multinomial Loit/Probit

Longitudinal data

  • Fixed Effects

  • Random Effects

  • Two-way Fixed Effects

    • Difference-in-Differences

Time Series Analysis

  • Univariate data

    • Arima / ArimaX
  • Multivariate Time Series

    • VAR

    • Error Correction …

Causality

  • Treatment effect models

    • Instrumental Variables

    • Difference-in-Differences

    • Regression Discontinuity

    • Propensity Score Matching

Predictions

  • Margins

  • Counterfactuals