16.9 In Section 16.6.3 we estimated a Poisson regression explaining the number of Olympic Games medals won by various countries as a function of the logarithms of population and gross domestic product (in 1995 dollars). The estimated coefficienr-.

are in Table 16.6.

(a) In 1988 Australia had GDP:3.0E* 11 and a population of 16.5 million_ Predict the number of medals that Australia would win. They did win 14 medals-

Calculate the probability that Australia would win 10 medals or more.

(b) In 1988 Canada had GDP:5.19E* 11 and a population of 26.9 million.

Predict the number of medals that Canada would win. They did win 10 medals-

Calculate the probability that they would win 15 medals or less.

(c) Use the combined data on years 7992 and 1996 to estimate the model explaining medals won as a function of the logarithms of population and gross domestic product. Compare these estimates to those in Table 16.6.

(d) In addition to population and GDP, the file olympics.dat contains a dumml vaiable(SOVIEZ) to indicate that a country was part of the former SovietUnion.

The dummy variable HOSTindicates the country hosting the Olympic Games.

Using again the combined data for 1992 and 1996, estimate the Poisson regression model that adds these two variables to the specification. Discuss the results. Are the signs what you expected? Are the added variables statisticalll'

significant?

(e) A variable similar to SOVIET is PLANNED, which includes nonmarket, typically communist countries. Use this variable instead of SOVIET and repeat (d).

Which model do you prefer, the one with SOVIET or the one with pI,ANNED?

whv?

(f) In 2000t,h eG DP (in 1995U S$ ) of Australiwa as3 .22224*E 11a ndt hato f Canada was 6.41256E+ i1. The Australian population in 2000 was 19.071 million, and that of Canada was 30.689 million. Using these figures, predict the number of medals won by Canada and Australia based on the estimates in part (e). Note that the 2000 games were held in Sydney, Australia. In 2000, Australia won 58 medals and Canada won 14. How close were your predictions?