A. Colin Cameron and Per Johansson, "Count Data Regression Using Series
Expansion:  With Applications", Journal of Applied Econometrics, Vol. 12,
No. 3, 1997, pp. 203-224.

*---- SUMMARY

We have provided Gauss programs for the computation of Poisson
Polynomial of order p (p =1,...,6) computed in the paper.

Also the Data on Takeover Bids (Section 4.1) and 
Health Service Utilization (Section 4.2) are provided.


*---- PROGRAMS (Program.zip)

1. Maincode to perform fast simulated annealing
   (FSA) 

   FSA2.SRC     Gauss program containing utilty progamms
		and programs for FSA.
   FSA2.EXT     External variable declarations for common globals 
   FSA2.DEC     External variable declarations for common globals
   FSA2.LCG     Library file for the procedures in FSA2.SRC

The first three file should be copied in to the SRC library in Gauss.
The last file should be copied into the lib library in Gauss.

2. Procedures to perform FSA and maximum likelihood estimation of the
PP1 - PP6 models the loglikelihood (l), gradient (g) and Hessian (H)
Also, supplemented are Procedures to calculated expectation (E),
variance (V), Ratio (R), sampling distribution (f) and pdf's from the 
PPp disstribution (psnp) and marginal effects (Sb)

PP1.SRC      l, g, H, E, V, R, f, psnp and Sb  
PP2.SRC      l, g, H, E, V, R, f, psnp and Sb
PP3.SRC      l, g, H, E, V, R, f, psnp and Sb
PP4.SRC      l, g, H, E, V, R, f, psnp and Sb
PP5.SRC      l, g, H, E, V, R, f, psnp and Sb
PP6.SRC      l, g

To perform ML estimation the Gauss Maxlik Procedures must be available.

3. An example file based on the the Takeover bid data.
Snp1.prg
This is en example file how one can use the programs above. To perform
FSA and ML of the PP1 model.
  
*---- DATA  (Data.zip)  These files are in DOS format.

4. Takeover bid data 
take.dat                                
Nr of observations     126                    
nr of variables         9                   
Variables in the following order:
  NUMBIDS   Count (Dependent Variable)
  LEGLREST  Legal Restructuring
  REALREST  Real Restructuring
  FINREST   Financial Restructuring
  WHTKNGHT  White Knight
  BIDPREM   Bid Premium
  INSTHOLD  Institutional Holdings
  SIZE      Size measured in billions
  REGULATN  Regulation
In addition you need to create the variable
  SIZESQ    Size Squared (SIZE*SIZE)
        

5. The Health Data-set (Ascci)
health.dat
Nr of observations    5190 
nr of variables         12
Variables in the following order:        
NONDOCCO SEX  AGE INCOME LEVYPLUS FREEPOOR 
FREEREPA ILLNESS ACTDAYS HSCORE CHCOND1 CHCOND2 


*---- DOCUMENTATION FOR THE TAKEOVER BID DATA SET

The original data are from Sanjiv Jaggia and Satish Thosar, 1993,
"Multiple Bids as a Consequence of Target Management Resistance",
Review of Quantitative Finance and Accounting, 447-457.

A related study that uses similar, but not exactly the same data,
is Sanjiv Jaggia and Satish Thosar, 1995,
"Contested Tender Offers: An Estimate of the Hazard Function",
Journal of Business and Economic Statistics, 113-119.
This related study looks at 152 firms, including 25 that were 
still not taken over 52 weeks after the initial offer. 
These different data are available at the JBES web-site.


*---- DOCUMENTATION FOR THE HEALTH DATA SET

The original data are a random sample of single Australians over 18
years of age.  Specifically, there are 40,650 people in the AHS 1977-78
study. 

For study description, codebook and questionnaire see
"Australian Health Survey, 1977-78: User's guide for the 
Machine-Readable Data File", SSDA Study No. 196,     
Social Sciences Data Archive, The Australian National 
University, Canberra, ACT 0200 AUSTRALIA
This can be purchased for AUD$17, at the above address.

Or see "Australian Health Survey, 1977-78: Sample File on Magnetic 
Tape", Information Paper, Catalogue Number 4324.0, 
Australian Bureau of Statistics, Canberra.

We first restrict attention to the 7,391 people in head-only households.
(i.e. Structure of Income Unit is head only). By restriction to singles
only this group has relatively more young and old people and relatively
fewer middle-age people than the population as a whole. 
Further restriction to only people over 18 years of age and with
non-missing data on all variables used in this and related analyses 
reduces the sample in the extracted data set to 5,190. 

The extracted data (or closely related extracted data sets) are 
used in four other studies in addition to this JAE paper:

A. Colin Cameron and Pravin K. Trivedi (1986), "Econometric Models 
Based on Applications of Some Estimators and Tests", Journal of 
Applied Econometrics, 1, 29-53.

A. Colin Cameron, Pravin K. Trivedi, Frank Milne and John R. 
Piggott (1988), "A Microeconometric Model of the Demand for 
Health Care and Health Insurance in Australia", Review of 
Economic Studies, 55, 85-106. 

A. Colin Cameron and Pravin K. Trivedi (1990), "The Role of Income
and Risk in the Choice of Health Insurance: Evidence from Australia", 
Journal of Public Economics, 45, 1-28.

A. Colin Cameron and Frank Windmeijer (1996). "R-Squared Measures for
Count Data Regression Models With Applications to Health Care
Utilization", Journal of Business and Economic Statistics, 14, 209-220. 

These papers include summaries of the data.  The second and third
papers include summaries of institutional arrangements of health
insurance in Australia.