EC 510/410 Econometric Analysis of Panel Data

Fall 2015, 4:40-6:30pm TTH (TBA)
Prof. K.-P. Lin (CH 241G, 725-3931)
Office Hours: 3:30-4:30 TTH & by appointment

Suggested Readings on R

• CRAN Task View: Econometrics
• Grant V. Farnsworth, Econometrics in R, 2008.
• Christian Kleiber and Achim Zeileis, Applied Econometrics with R, Springer-Verlag, New York, 2008.
• Yves Croissant and Giovanni Millo, Panel Data Econometrics in R: The plm Package, Journal of Statistical Software 27:2, 2008.
• Giovanni Millo and Gianfranco Piras, splm: Spatial Panel Data Models in R, Journal of Statistical Software 47:1, 2012.

Example 1: Investment Function

I_it = α_i + βF_it + γC_it + ε_it

where
i = 10 firms: GM, CH, GE, WE, US, AF, DM, GY, UN, IBM.
t = 20 years: 1935-1954.
I_it = Gross investment.
F_it = Market value.
C_it = Value of the stock of plant and equipment.
e_it = Error term.

Data of above 3 variables for 10 companies are available:
General Motors (GM)
Chrysler (CH)
General Electric (GE)
Westinghouse (WE)
U. S. Steel (US)
Atlantic Refining (AF)
Diamond Match (DM)
Goodyear (GY)
Union Oil (UN)
IBM
Alternatively, a combined (stacked) data is given here.

Example 2: International Comparison of Economic Growth

y_it = α_t + ρy_it-1 + β₁ln(s_i) + β₂ln(n_i+g+δ) + β₃COM_i + β₄OPEC_i + ε_it

where
y_it = Real per capita GDP
s_i = Average saving rate (over 1960-1985)
n_i = Average population growth rate (over 1960-1985)
g+δ = 5%
COM_i = 1 if communist, 0 otherwise
OPEC_i =1 if OPEC, 0 otherwise

Suggested Readings:
G. Mankiw, D. Romer, and D. Weil, "A Contribution to the Empirics of Economic Growth," Quarterly Journal of Economics 107, 1992, 407-437.
R. Summers and A. Heston, "The Penn World Table (Mark 5): An Expanded Set of International Comparisons, 1950-1988," Quarterly Journal of Economics 106, 1991, 327-368.

Example 3: Returns to Schooling

EXP =Work experience
WKS =Weeks worked
OCC =Occupation, 1 if blue collar
IND =1 if manufacturing industry
SOUTH =1 if resides in south
SMSA =1 if resides in a city (SMSA)
MS =1 if married
FEM =1 if female
UNION =1 if wage set by union contract
ED =Years of education
BLK =1 if individual is black
LWAGE=Log of wage

Suggested Readings:
Cornwell, C. and Rupert, P., "Efficient Estimation with Panel Data: An Empirical Comparison of Instrumental Variable Estimators," Journal of Applied Econometrics, 3, 1988, pp. 149-155.

Example 4: Wage Equation

The data is available in two parts:

• Part 1: Time-Variant Data (17,919 obs.) (Data)
  id = Person id (ranging from 1 to 2178)
  ed = Education
  lwage = Log of hourly wage
  pexp = Potential experience
  trend = Time trend

• Part 2: Time-Invariant Data (2178 individuals) (Data)
  ability = Time invariant ability
  medu = Mother’s education
  fedu= Father’s education
  d = Dummy variable for residence in a broken home
  siblings = Number of siblings
  

Example 5: U.S. Productivity

Munnell Productivity Data, 48 Continental U.S. States, 17 years from 1970 to 1986 (Data):

ST_ABB=State abbreviation
Region = 1, ..., 9
YR = Year: 1970, . . . ,1986
PCAP =Public capital
HWY =Highway capital
WATER =Water utility capital
UTIL =Utility capital
PC =Private capital
GSP =Gross state product
EMP =Employment

Homework

1. Using data of Example 4, consider the following wage equation:

   lwage_it = α₀ + α₁ability_i + α₂medu_i + α₃fedu_i + α₄d_i + α₅siblings_i + β₁ed_it + β₂pexp_it + ε_it

   Notice that all the α coefficients are associated with time-invariant cross section data, while β are with time-variant panel data series. Formulate, estimate, and compare the pooled or population-averaged based on OLS and OLS with panel-robust standard errors, respectively. In addition to pooled model, three different variable transformations should be considered and compared: (1) first-difference, (2) between (or group means), and (3) within (or deviations from group means). Note: not all coefficients can be estimated for all models. Why?

   If you are interested in the original paper, read this, but we are not attempting to replicate their results (see also Joshua C. C. Chan, "Replication of the Results in 'Learning about Heterogeneity in Returns to Schooling'", Journal of Applied Econometrics, Vol. 20. No. 3, 2005, pp. 439-443.)

2. Continuing on Homework 1, formulate, estimate, and compare the fixed-effects and random-effects panel data models based on OLS and OLS with panel-robust standard errors, respectively.

   Setup and perform hypothesis testings to choose a proper panel data model: (1) pool or not to pool? (2) fixed-effects or random-effects?