A financial analyst wishes to test the relationship between GNP and foreign earnings from developing countries. Theoretically, growth should be higher for those countries with a high foreign income and is external to the local economy. The two sources most readily measured are foreign direct investment and earning from exports. Use Appendix A to answer the following questions. GNP is in billion dollars while all other variables are expressed in million dollars.

a) Discuss the characteristic of the data by making reference to the lower subscript.

b) What signs would you expect on Beta2 and Beta3 ? Why?

c) Report the least squares estimates of the coefficients and corresponding p- values.

Em 10.0 copyright 1984-2008 Statistics/Data Analysis statacorp 4905 Lakeway Drive special Edition college Station, Texas 77845 USA 800-STATA-PC http:/ /www. stata. com 979-696-4600 statalstata. com 979-696-4601 ( fax) 20-user Stata for Windows (network) perpetual license: Serial number: 81910524292 Licensed to: school of Economics and Finance university of Western sydent Notes: (/m# option or -set memory-) 10. 00 MB allocated to data (/v#option or -set maxvar-) 5000 maximum variables 3. New update available; type -update all- insheet using "E: \\My Documents \\Teaching\\FM2009\\Exercises\\sessi on\\ForeignEarnings.txt" (4 vars, 25 obs) tsset obs time variable: obs, 1 to 25 delta: 1 unit summarize gnp directforeigninvestment exports variable Obs Mean Std. Dev. Min Max gnp 25 124.2 111. 4342 11 451 direct fore-t 25 142 . 64 98. 36026 22 38 8 exports 256.72 189. 4172 723 regress gnp directforeigninvestment exports source 55 di MS Number of obs 25 F( 2, 22) 65.90 Model 255393.33 2 127696.665 Prob> F 0. 0000 Residual 42628. 6696 22 1937 .6668 R-squared 0. 8570 Adj R-squared = 0.8440 Total ZZOB6Z 24 12417.5833 Root MSE 44.019 gnp coef. std. Err. t Polt1 [958 conf. Interval] directforest -. 001395 . 1593158 -0. 01 0.993 -.3317957 3290056 exports . 545196 . 0827292 6.59 0.000 . 3736261 . 7167659 cons -15.56372 15.94247 -0.98 0. 340 -48. 62639 17.49894estat hettest Breusch-Pagan / Cook-Weisberg test for heteroskedasticity Ho: constant variance Variables: fitted values of gnp chi 2 (1) 11. 94 Prob > chi 2 0.0005 es tat vi f variable VIF 1/VIF di rectforest 3.04 0. 328784 exports 3.04 D. 328784 Mean VIF 3.04 es tat hettest, ii d Breusch-Pagan / Cook-Weisberg test for heteroskedasticity Ho: Constant variance variables: fitted values of gnp chi 2(1) 8.46 Prob > chi2 0.0036 estat dwatson Durbin-Watson d-statistic( 3, 25) = 1. 215631 correlate directforeigninvestment exports (obs=2 5) direct t exports directforest 1.0000 exports 0. 8193 1.0000 Dickey-Fuller test for unit root Number of obs Interpolated Dickey-Fuller -.- Tes H Critical 5 Cntlcal 10%% Critical Statistic Value Value -5.137 -4 . 159 -3.501 -3.182 MacKennon approximate p-value for Z() 0.0001 predict error, .;b $2;test error Skewness/Kurtoals tests for Normality JOInt-- Van able Obs PriSkewmessl Pr(KU0615) adj ch12121 Prob> Chl2 error 50 0.0D07 0. IT70 10.90 0.000, summarize error, detail Linear prediction Percentiles Smallest 8.39421 8.39421 S! 14.88495 8.39421 10% 25.2339 14.88495 Obs 25% 50.79673 14.88495 Sum of Wet. 50 501 98.7405 Mean 124.2 Largest Sid. Dev. 102 0991 75! 171.4711 353.7814 90! JS.0458 151.7814 Variance 10424.22 95! 351.7814 178.0717 Skewness 1 26024 99! 178.0717 378.0717 Kurtosis 3. 700641