The HPPANEL Procedure

Getting Started: HPPANEL Procedure

The following statements use the cost function data from Greene (1990) to estimate the variance components model. The variable Production is the log of output in millions of kilowatt-hours, and the variable Cost is the log of cost in millions of dollars. See Greene (1990) for details.

data greene;
   input firm year production cost @@;
datalines;
1 1955   5.36598   1.14867  1 1960   6.03787   1.45185
1 1965   6.37673   1.52257  1 1970   6.93245   1.76627
2 1955   6.54535   1.35041  2 1960   6.69827   1.71109
2 1965   7.40245   2.09519  2 1970   7.82644   2.39480
3 1955   8.07153   2.94628  3 1960   8.47679   3.25967

   ... more lines ...

You decide to fit the following model to the data,

$C_{it}= \mr{Intercept} + {\bbeta }P_{it}+v_{i}+e_{t}+{\epsilon }_{it} \; \; \text {for }i=1, {\ldots }, \mi{N}\text { and }t=1, {\ldots }, \mi{T}$

where ${C_{it}}$ and ${P_{it}}$ represent the cost and production; and ${v_{i}}$ , ${e_{t}}$ , and ${{\epsilon }_{it}}$ are the cross-sectional, time series, and error variance components, respectively.

If you assume that the time and cross-sectional effects are random, four possible estimators are left for the variance components. The following statements choose the Fuller-Battese method to fit this model:

proc hppanel data=greene;
   model cost = production / rantwo vcomp = fb;
   id firm year;
   performance nodes=0 nthreads=2;
run;

The output of the HPPANEL procedure is shown in Figure 7.1.

Figure 7.1: Two-Way Random Effects Results

The HPPANEL Procedure

Model Information
Data Source	GREENE
Response Variable	cost
Model	RANTWO
Variance Component	FULLER
Execution Mode	Single-Machine

Fit Statistics
Sum of Squared Error	0.34808
Degree of Freedom	22.00000
Mean Squared Error	0.01582
Root Mean Squared Error	0.12579
R-Square	0.81362

Variance Component Estimates
Variance Component for Cross Sections	0.0469
Variance Component for Time Series	0.00906
Variance Component for Error	0.00875

Parameter Estimates
Parameter	DF	Estimate	Standard Error	t Value	Pr > \|t\|
Intercept	1	-2.99992	0.64778	-4.63	<.0001
production	1	0.74660	0.07618	9.80	<.0001

Printed first is the model description, which reports the method used for estimation and the method used for estimating error components. Printed next is the fit statistics table, and then the variance components estimates. Finally, the table of regression parameter estimates shows the estimates, standard errors, and t tests.