SOURCE Josef C BRada and Ronald L Graves The slowdown in Soc

SOURCE: Josef C. BRada and Ronald L. Graves, \"The slowdown in Societ Defense Expenditures\" Southern Economic Journal VOL. 54, No 4, p. 974
Datafile = DEFEND9

A.) The authoris expected positive signs for ALL the slope coefficients of both equations. Test these hypothses at the 5% level.

B.) Use our four specification criteria to determine whether SP is an irrelevant varaible. Explain your reasoning.

C.) Test both eqations for positive first order serial correlation. Does the high probability of serial correlation cause you to reconsider your answer to part B? Explain.

D.) Somone might argue that because the DW statistic improved when LnSP was added that the serial correlation was impure and that GLS was not called for. Do you agree with this conclusion? Why or why not?

E.) If we run a GLS version of Equation 9.26 we get Equation 9.28. Does this resulat cause you to reconsider your answer in part B? Exaplain..

YEAR	SDH	SDL	USD	SY	SFP	NR	NU
1960	31	23	200.54	232.3	7.03	415	1734
1961	34	26	204.12	245.3	6.07	445	1846
1962	338	29	207.72	254.5	3.90	485	1942
1963	39	31	206.86	251.7	2.97	531	2070
1964	42	34	207.41	279.4	1.40	580	2910
1965	43	35	185.42	296.8	1.87	598	4110
1966	44	36	203.19	311.9	4.10	674	4198
1967	47	39	241.21	326.3	4.90	1058	4338
1968	50	43	254.62	355.9	2.87	1662	4026
1969	52	43	254.62	355.9	2.87	1662	4026
1970	53	44	228.19	383.3	4.43	2047	5074
1971	54	45	206.80	398.2	3.77	3199	6282
1972	56	46	189.41	405.7	2.87	2298	7100
1973	58	48	169.27	435.2	4.30	2534	8522
1974	65	53	155.59	459.8	6.33	2614	9170
1975	65	53	155.59	459.8	6.33	2614	9170
1976	69	56	169.91	481.8	0.63	3219	9518
1977	70	56	170.94	497.4	2.23	4345	9806
1978	72	57	154.12	514.2	1.03	5097	9950
1979	75	56	156.80	516.1	0.17	6336	9945
1980	79	62	16.67	524.7	0.27	7451	9668
1981	83	63	169.55	536.1	0.47	7793	9628
1982	84	64	185.31	547.0	0.07	8031	101201
1983	90	67	211.5	578.9	1.63	9146	10630

14. In a 1988 article, Josef Brada and Ronald Graves built an interesting model of defense spending in the Soviet Union just before the breakup of that nation. 13 The authors felt sure that Soviet defense spending was a function of U.S. defense spending and Sovie GNP but were less sure about whether defense spending also was a function of the ratio of Soviet nuclear warheads to U.S. nuclear warheads. Using a double-log functional form, the authors estimated a number of alter- native specifications, including (standard errors in parentheses): In SDH 1.99 0.056lnUSD 0.969 nSY 0.057lnSPt (9.26) (0.074) t = 0.76 (0.065)(0.032) 14.98 1.80 N-25 (annual 1960-1984) R2-.979 DW=0.49 In SDH 2.88 0.105lnUSD 1.066lnSY (9.27) (0.038) 28.09 (0.073) t 1.44 N 25 (annual 1960-1984) R2 .977 DV -0.43 the CIA\'s \"high\" estimate of Soviet defense expendi- tures in year t (billions of 1970 rubles) where: SDH, USD SY Soviet GNP in year t (billions of 1970 rubles) U.S. defense expenditures in year t (billions of 1980 dollars) the ratio of the nurnber of USSR nuclear warheads (NR) to the number of U.S. nuclear warheads (NU) in year t SP, 13. Josef C. Brada and Ronald L Graves, \"The Slowdown in Soviet Defense Expenditures,\" Southern Economic Journal, Vol. 54, No. 4, pp. 969-984. In addition to the variables used in this exercise, Brada and Graves also provide data for SFP the rate of Soviet factor productivity in year t, which we include in Table 9.2 because we suggest exercises using SFP in the instructor\'s manual.

Solution