Practice Part

Empirical question. E4.4 of Stock and Watson a) Construct a scatterplot of average annual growth rate (Growth) on the average trade share (TradeShare). Does there appear to be a relationship between the variables?

-2

0

2

growth

4

6

8

scatter growth tradeshare

0

.5

1 tradeshare

1.5

2

It appears to be a positive relationship between growth and trade because of the positive slope of the line that would go through that scatterplot of data points. b) One country, Malta, has a trade share much larger than the other countries. Find Malta on the scatterplot. Does Malta look like an outlier? scatter

growth tradeshare, mlabel( country_name)

8

Korea, Republic of Taiwan, China

6

Malta

Cyprus Thailand Japan

2 -2

0

growth

4

Malaysia Portugal Greece NorwayIreland Mauritius Iceland Brazil Spain Italy Finland Austria Pakistan IsraelSri Lanka Belgium Dominican Republic Germany France Canada Ecuador Paraguay Netherlands Colombia Denmark Panama Australia Mexico United Kingdom Kenya India Sweden United States Costa Rica Chile Switzerland Philippines New Trinidad ZealandandPapua Tobago New Guinea Uruguay Guatemala Zimbabwe Bangladesh Honduras Argentina Jamaica South Africa Bolivia Togo Peru Sierra Leone ElSenegal Salvador Haiti Venezuela Ghana

Niger Zaire

0

.5

1 tradeshare

1.5

2

Malta does look like an outlier. We can also find it using Stata. If we define outlier as an observation that is 4 s.d. away from the mean, we have: egen trade_sd=sd(tradeshare) encode country_name, gen(country) gen difference=[abs(tradeshare-trade_sd)]/trade_sd table country if difference>=4, c(max difference) ------------------------country | max(differ~e) ----------+-------------Malta | 5.888421 -------------------------

Malta trade share is 5.89 s.d. away from the mean.

c) Using all observations, run a regression of Growth on TradeShare. What is the estimated slope? What is the estimated intercept? Use the regression to predict the growth rate for a country with trade share of 0.5 and with trade share equal to 1.0.

reg growth tradeshare Source | SS df MS -------------+-----------------------------Model | 28.4885066 1 28.4885066 Residual | 201.851551 63 3.20399287 -------------+-----------------------------Total | 230.340057 64 3.59906339

Number of obs F( 1, 63) Prob > F R-squared Adj R-squared Root MSE

= = = = = =

65 8.89 0.0041 0.1237 0.1098 1.79

-----------------------------------------------------------------------------growth | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------tradeshare | 2.306434 .773485 2.98 0.004 .7607473 3.85212 _cons | .6402653 .4899767 1.31 0.196 -.3388749 1.619405 ------------------------------------------------------------------------------

Estimated slope = 2.306434 Estimated intercept= 0.6402653 Trade share=0.5  Estimated Growth= 0.6402653+2.306434*0.5 = 1.7934823 Trade share=1.0  Estimated Growth= 0.6402653+2.306434*1 = 2.9466993

d) Estimate the same regression excluding the data from Malta. Answer the same questions in c reg

growth tradeshare if country_name!="Malta"

Source | SS df MS -------------+-----------------------------Model | 9.28031557 1 9.28031557 Residual | 198.527844 62 3.20206201 -------------+-----------------------------Total | 207.80816 63 3.29854222

Number of obs F( 1, 62) Prob > F R-squared Adj R-squared Root MSE

= = = = = =

64 2.90 0.0937 0.0447 0.0292 1.7894

-----------------------------------------------------------------------------growth | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------tradeshare | 1.680905 .9873624 1.70 0.094 -.2928046 3.654614 _cons | .9574107 .5803727 1.65 0.104 -.2027378 2.117559 ------------------------------------------------------------------------------

Estimated slope = 1.680905 Estimated intercept= 0.9574107 Trade share=0.5  Estimated Growth = 0.9574107+1.680905*0.5 = 1.7978632 Trade share=1.0  Estimated Growth = 0.9574107+1.680905*1 = 2.6383157

e) Where is Malta? Why is Malta trade share so large? Should Malta be included or excluded from the analysis?  

Malta is in Southern Europe, is an island in the Mediterranean Sea, south of Sicily (Italy). It is very small (316 sq km). Its location, in the Centre of the Mediterranean, gives it strategic importance. The economy is dependent on foreign trade (serving as a freight trans-shipment point), manufacturing (especially electronics and textiles) and tourism.

 

Malta’s trade share is very large because, first, when you are a small island you are likely to depend on foreign trade to get goods. Malta is an exporter of limestone. Secondly, Malta has an strategic geographic location. Malta should be excluded from the analysis otherwise we would be overestimating the effect of trade on growth. Malta is an exceptional case of trade share.