1 of 15

1

Webinar Series in Applied Quantitative Analysis - Updated

https://www.airp3-africa.org/applied-quantitative-analysis-webinar-series

Date

Topic

�February 29�March 7

Session One�Potential Outcomes and Omitted Variable Bias I (Theory) �Potential Outcomes and Omitted Variable Bias II (Application)

�March 21�March 28

Session Two�Difference-in-differences I (Theory)�Difference-in-differences II (Application)

�April 25�May 2

Session Three�Power analysis, clustering and sample size calculations I (Theory)�Power analysis, clustering and sample size calculations II (Application)

 

May 23�May 30

Session Four�Propensity score matching (Theory)�Propensity score matching (Application)

�June 20�June 27

Session Five�Fixed-effects I (Theory)�Fixed-effects II (Application)

�July 25�August 1

Session Six�Instrumental variables I (Theory)�Instrumental Variables II (Application)

 

August 22

August 29

Session Seven�Lagged dependent variables and the Arellano-Bond Estimator I (Theory)�Lagged dependent variables and the Arellano-Bond Estimator II (Application)

2 of 15

������ �Écoute de l'interprétation d'une langue �Windows | macOS�

2

�1. Dans les contrôles de votre réunion/webinaire, cliquez sur Interprétation .

2. Cliquez sur la langue que vous souhaitez entendre. (Nous aurons le français) Pas besoin de choisir l'anglais, c'est la langue de la salle Zoom principale

​

​

​

​

​

​

​

3. (Facultatif) Pour entendre uniquement la langue interprétée, cliquez sur Couper le son original.

Remarques:

  • Vous devez rejoindre l’audio de la réunion via l’audio/VoIP de votre ordinateur. Vous ne pouvez pas écouter l’interprétation linguistique si vous utilisez les fonctions audio de connexion ou d’appel téléphonique.
  • En tant que participant rejoignant une chaîne linguistique, vous pouvez retransmettre sur le canal audio principal canal si vous réactivez votre audio et parlez.

​

​

​

​

​

.

3 of 15

Listening to language interpretation� Windows | macOS

3

  1. In your meeting/webinar controls, click Interpretation .
  2. Click the language that you would like to hear. (We will have French) No need to choose English, that is the language in the main Zoom room

​

​

​

​

​

​

​

​

3. (Optional) To hear the interpreted language only, click Mute Original Audio.

Notes:

      • You must join the meeting audio through your computer audio/VoIP. You cannot listen to language interpretation if you use the dial-in or call me phone audio features.
      • As a participant joining a language channel, you can broadcast back into the main audio

channel if you unmute your audio and speak.

4 of 15

Dynamic Panel Data Estimation II - Application

Ashu Handa

Institute Fellow – AIR

Kenan Eminent Professor of Public Policy – UNC-CH

August 29, 2024

4

5 of 15

Recap of the main problem and the Blundell-Bond (1998) solution

Hit = β0 + β1(Hit-1) + β2(Xit) + β3(Xht) + (εit + µi)

Current period (t) height depends on lagged height (t-1)

Truly random error: not correlated

with Xs nor Hit-1

Health endowment of child: fixed over

time and correlated with Hi in each period

6 of 15

We have an endogeneity problem: Hit-1 is correlated with the error term!

Hit = β0 + β1(Hit-1) + β2(Xit) + β3(Xht) + (εit + µi)

We can solve this using IV. Instruments must satisfy two criteria

  1. Must be highly correlated with Hit-1;
  2. Must not directly affect Hit;

7 of 15

The Blundell-Bond (1998) system estimator

Hit = β0 + β1(Hit-1) + β2(Xit) + β3(Xht) + (εit + µi) (1)

Use (Hit-1 – Hit-2) as an instrument for Hit-1

Use Hit-2 as the instrument for (Hit-1 – Hit-2)

ΔHit = β1(ΔHit-1) + β2(ΔXit) + β3(ΔXht) + (Δεit) (3)

Levels equation with a lagged

differenced instrument

Differenced equation with a lagged

levels instrument

8 of 15

Hit = β0 + β1(Hit-1) + β2(Xit) + β3(Xht) + (εit + µi)

9 of 15

Weak instrument problem identified by Blundell-Bond

10 of 15

Differenced equation with

two period lagged instrument in levels

System estimator

11 of 15

System estimator

Other exogenous variables can serve as instruments. In this case we have used prices and long-run wealth.

12 of 15

xtdpd htz l.htz X1 Prices, dgmmiv(htz) lgmmiv(htz) div(Prices) liv(Lprices) artests(1) two vce(robust)

​

STATA command

‘dynamic panel data’ estimation

Dependent variable: htz�Lagged dependent variable as predictor: l.htz�(lag operator in STATA, must tsset the data)

Other exogenous variables

in the model

13 of 15

xtdpd htz l.htz X1 Prices, dgmmiv(htz) lgmmiv(htz) div(Prices) liv(Lprices) artests(1) two vce(robust)

​

STATA command

‘dynamic panel data’ estimation

‘GMM-style’ IV for the differenced equation

[This is the lagged dependent variable]

[STATA knows that this must be lagged two periods]

‘GMM style’ IV for the levels equation�[This is the lagged dependent variable]

[STATA knows this must be differenced]

14 of 15

xtdpd htz l.htz X1 Prices, dgmmiv(htz) lgmmiv(htz) div(Prices) liv(Lprices) artests(1) two vce(robust)

​

STATA command

‘dynamic panel data’ estimation

Other potential instruments for the differenced equation

They will be in levels or lagged levels, you must create these by hand

Other potential instruments for the levels equation

They will be in lags or lag differences, you must

create these by hand

15 of 15

xtdpd htz l.htz X1 Prices, dgmmiv(htz) lgmmiv(htz) div(Prices) liv(Lprices) artests(1) two vce(robust)

​

STATA command

‘dynamic panel data’ estimation

Other options, read about them by typing ‘help xtdpd’�I wont spend much time on them