ans_hw4_eco321_2010f

# ans_hw4_eco321_2010f - Economic Statistics II Solutions...

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Economic Statistics II Solutions - Homework Assignment 4 Use the data set usmacro.dta that contains the monthly US unemployment rate [%]: Stata's variable name unemp, from 1984m1 to 2010m3. Define U nempt to be the unemployment rate. Assume that the error term is homoskedastic. (a) Construct a graph of the historical data of the unemployment rate. FIGURE-1: Unemployment Rate 12 10 unemp 8 6 4 1985m1 1993m5 time 2001m9 2010m1 (b) Find the 1st- to 10th-autocorrelation of U nempt . TABLE 1: Autocorrelations Lag 1 2 3 4 5 6 7 8 9 10 U nemp .0593 -.1555 -.3307 -.0340 .2207 .1107 .2164 -.0444 -.3693 -.1720 1 (c) Run the regression of AR(1), AR(2), AR(3), and AR(4) model for U nempt . TABLE 2: Autoregressions of U nempt AR(1) U nempt-1 U nempt-2 U nempt-3 U nempt-4 .9687 (.0160) AR(2) 1.0471 (.0565) -.0809 (.0572) AR(3) 1.0569 (.0564) -.2245 (.0815) .1432 (.0571) AR(4) 1.0098 (.0540) -.1531 (.0782) -.1861 (.0788) .3233 (.0549) .0455 (.0967) .9289 311 37.843 const. R T SSR 2 .1878 (.0964) .9209 314 43.642 .2031 (.0984) .9205 313 43.264 .1494 (.1001) .9211 312 42.363 (d) Run the regression of AR(1), AR(2), AR(3), and AR(4) model for U nempt. TABLE 3: Autoregressions of U nempt AR(1) U nempt-1 U nempt-2 U nempt-3 U nempt-4 .0594 (.0160) AR(2) .0661 (.0565) -.1597 (.0572) AR(3) .0115 (.0538) -.1412 (.0532) -.3267 (.0542) AR(4) -.0096 (.0569) -.1501 (.0537) -.3325 (.0542) -.0417 (.0572) .0124 (.0199) .1275 310 37.456 const. R T SSR 2 .0054 (.0212) .00003 313 43.854 .0072 (.0210) .0225 312 42.653 .0102 (.0199) .1243 311 37.860 2 (e) Using BIC (SBIC in Stata), find which is the best model in (c) and (d), respectively. TABLE 4: BIC Lag 1 2 3 4 U nemp .9024 .9166 .9148 .8238* U nemp .9085 .8993 .8058* .8170 The AR(4) is the best model for U nemp, and the AR(3) for U nempt (f ) Conduct a DF test for the AR(1) of U nempt and U nempt at the 5% level, respectively. Are the AR(1) of U nempt and U nempt stationary? [Hint: Using estimates and standard errors from Stata, do DF test manually.] In DF test, H0 : 1 = 1 (nonstationary) vs H1 : 1 < 1 (stationary). U nempt : t-stat.= (.9687 - 1)/.01604 = -1.951 < -1.645 (Critical Value at 5%). We reject H0 . Thus, the AR(1) model is stationary. U nempt : t-stat.= (.05639 - 1)/.05653 = -16.69 < -1.645 (Critical Value at 5%). We reject H0 . Thus, the AR(1) model is stationary. Note: This is essentially a left-tail t-test. (g) Conduct an ADF test (with intercept only) for the AR(3) of U nempt at the 5% level. Is the AR(3) of U nempt stationary? In ADF test, H0 : nonstationary vs H1 : stationary Augmented Dickey-Fuller test for unit root Number of obs = 310 ----------- Z(t) has t-distribution ----------Test 1% Critical 5% Critical 10% Critical Statistic Value Value Value -----------------------------------------------------------------------------Z(t) -12.254 -2.339 -1.650 -1.284 -----------------------------------------------------------------------------p-value for Z(t) = 0.0000 Since the p-value is nearly 0%, which is less than 5%, we reject the null hypothesis. This suggests that the AR(3) model is stationary. Alternatively, since the test statistic, -12.254, is less than the critical value, -1.650, we reject H0 . Note that this is a left-tail test, but not t-test. (h) Using the AR(1) of U nempt , forecast (predict) U nempt of 2010m4. The actual value of U nempt of 2010m4 turned out to be 0.2 %. Find a forecast error (FE) and a root mean square forecast error (RMSFE) From the sample, we find Y2010m3 = -.2. ^ So, the forecast is: Y2010m4|2010m3 = .005447 + 3 .05939Y2010m3 = .005447 + (.05939)(-.2) = -.0064 [%] ^ The forecast error (FE) is Y2010m4|2010m3 - Y2010m3 = -.0064 - .2 = -.2064 [%] 2 = .2064 [%] RMSFE is (-.2064) Extra Credit: In (e), find BIC manually. U nemp AR(1): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(43.642/314) + (1 + 1)(ln(314)/314) AR(2): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(43.264/313) + (2 + 1)(ln(313)/313) AR(3): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(42.363/312) + (3 + 1)(ln(312)/312) AR(4): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(37.843/311) + (4 + 1)(ln(311)/311) U nemp AR(1): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(43.854/313) + (1 + 1)(ln(313)/313) AR(2): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(42.653/312) + (2 + 1)(ln(312)/312) AR(3): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(37.860/311) + (3 + 1)(ln(311)/311) AR(4): ln(SSR/T ) + (p + 1)(ln(T )/T ) = ln(37.456/310) + (4 + 1)(ln(310)/310) The AR(4) is the best model for U nemp, and the AR(3) for U nempt Appendix: Stata Outputs = -1.937 = -1.924 = -1.923 = -2.014 = -1.929 = -1.935 = -2.032 = -2.021 . tsset time , m o n t h l y time variable : delta : t i m e , 1984 m 1 1 month t o 2010 m 3 . generate (1 m i s s i n g . . graph7 dunemp = D . unemp value generated ) time , xlab ylab s (.) c(l) unemp corrgram dunemp , lags (10) -1 0 1 -1 0 1 LAG AC PAC Q P r o b >Q [ Autocorrelation ] [ Partial Autocor ] - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1 0.0593 0.0594 1.1161 0.2908 | | 2 -0.1555 -0.1597 8.8109 0.0122 -| -| 3 -0.3307 -0.3267 43.708 0.0000 --| --| 4 -0.0340 -0.0417 44.077 0.0000 | | 5 0.2207 0.1463 59.723 0.0000 |- |- 6 0.1107 -0.0117 63.672 0.0000 | | 7 0.2164 0.2848 78.809 0.0000 |- |-- 8 -0.0444 0.1036 79.448 0.0000 | | 9 -0.3693 -0.3596 123.81 0.0000 --| --| 10 -0.1720 -0.0863 133.47 0.0000 -| | . regress unemp L . unemp N u m b e r of obs F( 1, 312) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = 314 = 3647.34 = 0.0000 = 0.9212 = 0.9209 = .374 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | 510.18639 1 510.18639 Residual | 43.6422777 312 .139879095 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 553.828668 313 1.76942066 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | L1 . | .9687331 .0160404 60.39 0.000 .937172 1.000294 4 | _ cons | .1877781 .0963853 1.95 0.052 - . 0018693 .3774254 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . regress u ne m p L (1 / 2 ) . un e m p N u m b e r of obs F( 2, 310) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = 313 = 1806.14 = 0.0000 = 0.9210 = 0.9205 = .37358 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | 504.131065 2 252.065533 Residual | 43.2636386 310 .139560124 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 547.394704 312 1.7544702 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | L1 . | 1.047084 .0565519 18.52 0.000 .93581 1.158358 L2 . | - . 0808716 .0571968 -1.41 0.158 - . 1934147 .0316715 | _ cons | .2031179 .0984006 2.06 0.040 .0095003 .3967355 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . regress u ne m p L (1 / 3 ) . un e m p N u m b e r of obs F( 3, 308) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = 312 = 1211.74 = 0.0000 = 0.9219 = 0.9211 = .37087 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | 499.995374 3 166.665125 Residual | 42.3630612 308 .137542406 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 542.358435 311 1.7439178 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | L1 . | 1.056944 .056385 18.75 0.000 .9459951 1.167892 L2 . | - . 2244753 .0814709 -2.76 0.006 - . 3847853 - . 0641653 L3 . | .1431798 .0570996 2.51 0.013 .0308251 .2555345 | _ cons | .1493853 .1001379 1.49 0.137 - . 0476556 .3464263 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . regress u ne m p L (1 / 4 ) . un e m p N u m b e r of obs F( 4, 306) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = 311 = 1013.69 = 0.0000 = 0.9298 = 0.9289 = .35167 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | 501.451192 4 125.362798 Residual | 37.8427685 306 .123669178 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 539.293961 310 1.73965794 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - unemp | L1 . | 1.009807 .0540319 18.69 0.000 .9034856 1.116128 L2 . | - . 1530716 .0782423 -1.96 0.051 - . 3070327 .0008895 L3 . | - . 1861212 .0787774 -2.36 0.019 - . 3411352 - . 0311072 L4 . | .3233378 .0549669 5.88 0.000 .2151768 .4314988 | _ cons | .0454608 .0967223 0.47 0.639 - . 1448641 .2357857 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . regress dunemp L . dunemp N u m b e r of obs F( 1, 311) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = = = = = = 313 1.10 0.2943 0.0035 0.0003 .37551 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | .155637642 1 .155637642 Residual | 43.8540134 311 .141009689 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 44.009651 312 .141056574 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp | L1 . | .0593856 .056526 1.05 0.294 - . 0518362 .1706075 | _ cons | .0054472 .0212272 0.26 0.798 - . 0363198 .0472143 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . regress dunemp L (1 / 2 ) . d u n e m p N u m b e r of obs F( 2, 309) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = = = = = = 312 4.57 0.0110 0.0288 0.0225 .37153 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | 1.26286489 2 .631432445 Residual | 42.6530034 309 .13803561 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 43.9158683 311 .14120858 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 5 | | .0661495 .0561277 1.18 0.239 - . 0442913 .1765904 | - . 1597355 .0560497 -2.85 0.005 - . 2700229 - . 0494481 | _ cons | .0072283 .0210391 0.34 0.731 - . 0341697 .0486262 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp L1 . L2 . . regress dunemp L (1 / 3 ) . d u n e m p N u m b e r of obs F( 3, 307) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = = = = = = 311 15.67 0.0000 0.1328 0.1243 .35117 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | 5.79835069 3 1.93278356 Residual | 37.8599154 307 .1233222 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 43.6582661 310 .140833116 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp | L1 . | .0114654 .0537723 0.21 0.831 - . 0943436 .1172744 L2 . | - . 1412366 .0532115 -2.65 0.008 - . 2459419 - . 0365313 L3 . | - . 3266663 .0541591 -6.03 0.000 - . 4332363 - . 2200963 | _ cons | .0102198 .0199211 0.51 0.608 - . 0289794 .0494189 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . regress dunemp L (1 / 4 ) . d u n e m p N u m b e r of obs F( 4, 305) Prob > F R-s q u a r e d A d j R-s q u a r e d Root MSE = = = = = = 310 12.29 0.0000 0.1388 0.1275 .35044 Source | SS df MS - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Model | 6.03518011 4 1.50879503 Residual | 37.45579 305 .122805869 - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Total | 43.4909701 309 .140747476 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp | Coef . Std . Err . t P >| t | [95% C o n f . I n t e r v a l ] - - - - - - -+ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - dunemp | L1 . | - . 0096228 .0569536 -0.17 0.866 - . 1216946 .102449 L2 . | - . 1501377 .0536946 -2.80 0.005 - . 2557965 -.044479 L3 . | - . 3324914 .05419 -6.14 0.000 -.439125 - . 2258578 L4 . | - . 0416683 .0571849 -0.73 0.467 - . 1541952 .0708585 | _ cons | .0124501 .0199181 0.63 0.532 - . 0267441 .0516443 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . varsoc unemp , maxlag (4) S e l e c t i o n -o r d e r c r i t e r i a Sample : 1984 m 5 - 2010 m 3 N u m b e r of obs = 311 + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -+ | lag | LL LR df p FPE AIC HQIC SBIC | | - -+ - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - -- - - - - -- - - -- - - -| | 0 | -526.888 1.74525 3.39478 3.39958 3.4068 | | 1 | -134.665 784.45 1 0.000 .140999 .878872 .888485 .902922 | | 2 | -133.828 1.6741 1 0.196 .141147 .87992 .89434 .915996 | | 3 | -130.411 6.8333 1 0.009 .13897 .864379 .883605 .912479 | | 4 | -113.752 33.318 * 1 0.000 .125657 * .763678 * .787711 * .823804 * | + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -+ Endogenous : unemp Exogenous : _ cons . varsoc dunemp , maxlag (4) S e l e c t i o n -o r d e r c r i t e r i a Sample : 1984 m 6 - 2010 m 3 N u m b e r of obs = 310 + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -+ | lag | LL LR df p FPE AIC HQIC SBIC | | - -+ - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - -- - - - - -- - - -- - - -| | 0 | -135.448 .141202 .88031 .885128 .892363 | | 1 | -135.076 .74438 1 0.388 .141775 .88436 .893997 .908467 | | 2 | -130.637 8.8767 1 0.003 .138664 .862177 .876633 .898338 | | 3 | -112.562 36.151 * 1 0.000 .1242 * .752012 * .771286 * .800226 * | | 4 | -112.292 .53918 1 0.463 .124787 .756724 .780816 .816991 | + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -+ Endogenous : dunemp Exogenous : _ cons . dfuller dunemp , drift lags (3) test for unit root Number of obs = 310 Augmented D i c k e y -F u l l e r - - - - - - Z ( t ) h a s t-d i s t r i b u t i o n - - - - - - - - - - - - - - - - Test 1% C r i t i c a l 5% C r i t i c a l 10% C r i t i c a l Statistic Value Value Value - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Z(t) -12.254 -2.339 -1.650 -1.284 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - p-v a l u e f o r Z ( t ) = 0 . 0 0 0 0 . 6 ...
View Full Document

## This note was uploaded on 03/08/2012 for the course ECONOMICS 321 taught by Professor Hiroshimorita during the Spring '11 term at CUNY Hunter.

Ask a homework question - tutors are online