ABC Monday February 18 00:14:59 2019 Page 1

___ ____ ____ ____ ____(R)

/__ / ____/ / ____/
___/ / /___/ / /___/
Statistics/Data Analysis

User: ABC
Project: ABC

___ ____ ____ ____ ____ (R)

/__ / ____/ / ____/
___/ / /___/ / /___/ 16.0 Copyright 1985-2019 StataCorp LLC
Statistics/Data Analysis StataCorp
4905 Lakeway Drive
MP - Parallel Edition College Station, Texas 77845 USA
800-STATA-PC http://www.stata.com
979-696-4600 [email protected]
979-696-4601 (fax)

Single-user 2-core Stata network license expires 20 Aug 2019:

Serial number: 501609213901
Licensed to: Anhtony Charca
ACharca's Econometric Advisor Group

Notes:
1. Unicode is supported; see help unicode_advice.
2. More than 2 billion observations are allowed; see help obs_advice.
3. Maximum number of variables is set to 5000; see help set_maxvar.

1 . import excel "C:\Users\dangn\Desktop\SOS.xlsx", sheet("Sheet1") firstrow

(9 vars, 315 obs)

2 . encode Country, gen(COUNTRY)

3 . xtset COUNTRY Year

panel variable: COUNTRY (strongly balanced)
time variable: Year, 2001 to 2021
delta: 1 unit

4 . sum GDP NetFDI RIR Reserves CA LIR Covid

Variable Obs Mean Std. Dev. Min Max

GDP 315 2075.397 4499.895 32.68 22996.1

NetFDI 315 -13.1681 50.80397 -344.331 177.277
RIR 315 5.787484 9.003918 -20.5 48.34
Reserves 315 19.55524 74.63692 -443.625 479.5533
CA 315 -18.17763 152.3456 -821.645 420.5685

LIR 315 11.06029 10.46254 2.23 67.08

Covid 315 .0952381 .2940106 0 1

5 . corr GDP NetFDI RIR Reserves CA LIR Covid

(obs=315)

GDP NetFDI RIR Reserves CA LIR Covid

GDP 1.0000
NetFDI -0.1298 1.0000
RIR -0.0880 -0.0892 1.0000
Reserves 0.0916 -0.4378 -0.0699 1.0000
CA -0.6142 -0.3012 -0.0083 0.2183 1.0000
LIR -0.1710 -0.0930 0.8501 -0.0597 0.0189 1.0000
Covid 0.0842 -0.0215 -0.0694 0.0336 -0.0006 -0.0953 1.0000
 ABC Monday February 18 00:14:59 2019 Page 2

6 . xtunitroot fisher GDP , dfuller lag (0)

Fisher-type unit-root test for GDP

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 21

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 6.7574 1.0000

Inverse normal Z 4.9445 1.0000
Inverse logit t(74) L* 5.3470 1.0000
Modified inv. chi-squared Pm -3.0006 0.9987

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

7 . xtunitroot fisher d.GDP , dfuller lag (0)

(15 missing values generated)

Fisher-type unit-root test for D.GDP

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 20

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 180.6685 0.0000

Inverse normal Z -10.3366 0.0000
Inverse logit t(79) L* -12.8728 0.0000
Modified inv. chi-squared Pm 19.4512 0.0000

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

8 . xtunitroot fisher NetFDI , dfuller lag (0)

Fisher-type unit-root test for NetFDI

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 21
 ABC Monday February 18 00:14:59 2019 Page 3

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 37.2535 0.1699

Inverse normal Z -0.8489 0.1980
Inverse logit t(79) L* -0.8554 0.1975
Modified inv. chi-squared Pm 0.9364 0.1745

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

9 . xtunitroot fisher d.NetFDI , dfuller lag (0)

(15 missing values generated)

Fisher-type unit-root test for D.NetFDI

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 20

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 391.9233 0.0000

Inverse normal Z -17.0382 0.0000
Inverse logit t(79) L* -28.0730 0.0000
Modified inv. chi-squared Pm 46.7241 0.0000

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

10 . xtunitroot fisher RIR ,dfuller lag (0)

Fisher-type unit-root test for RIR

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 21

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 152.7930 0.0000

Inverse normal Z -8.1469 0.0000
Inverse logit t(79) L* -10.4986 0.0000
Modified inv. chi-squared Pm 15.8525 0.0000

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.
 ABC Monday February 18 00:14:59 2019 Page 4

11 . xtunitroot fisher Reserves , dfuller lag (0)

Fisher-type unit-root test for Reserves

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 21

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 128.1871 0.0000

Inverse normal Z -7.5018 0.0000
Inverse logit t(79) L* -8.9259 0.0000
Modified inv. chi-squared Pm 12.6759 0.0000

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

12 . xtunitroot fisher CA , dfuller lag (0)

Fisher-type unit-root test for CA

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 21

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 30.9642 0.4172

Inverse normal Z -0.6382 0.2617
Inverse logit t(79) L* -0.6211 0.2682
Modified inv. chi-squared Pm 0.1245 0.4505

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

13 . xtunitroot fisher d.CA , dfuller lag (0)

(15 missing values generated)

Fisher-type unit-root test for D.CA

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 20
 ABC Monday February 18 00:14:59 2019 Page 5

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 181.4921 0.0000

Inverse normal Z -10.5957 0.0000
Inverse logit t(79) L* -12.9622 0.0000
Modified inv. chi-squared Pm 19.5575 0.0000

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

14 . xtunitroot fisher Covid , dfuller lag (0)

Fisher-type unit-root test for Covid

Based on augmented Dickey-Fuller tests

Ho: All panels contain unit roots Number of panels = 15

Ha: At least one panel is stationary Number of periods = 21

AR parameter: Panel-specific Asymptotics: T -> Infinity

Panel means: Included
Time trend: Not included
Drift term: Not included ADF regressions: 0 lags

Statistic p-value

Inverse chi-squared(30) P 1.9955 1.0000

Inverse normal Z 5.8840 1.0000
Inverse logit t(79) L* 5.7529 1.0000
Modified inv. chi-squared Pm -3.6154 0.9999

P statistic requires number of panels to be finite.

Other statistics are suitable for finite or infinite number of panels.

15 . reg d.GDP d.NetFDI RIR Reserves d.CA LIR Covid

Source SS df MS Number of obs = 300

F(6, 293) = 17.43
Model 8596575.81 6 1432762.64 Prob > F = 0.0000
Residual 24079948.1 293 82184.1231 R-squared = 0.2631
Adj R-squared = 0.2480
Total 32676523.9 299 109286.033 Root MSE = 286.68

D.GDP Coef. Std. Err. t P>|t| [95% Conf. Interval]

NetFDI
D1. -.9210913 .3511653 -2.62 0.009 -1.612217 -.2299652

RIR 1.267604 3.473379 0.36 0.715 -5.568331 8.103538

Reserves 1.772065 .2213975 8.00 0.000 1.336334 2.207796

CA
D1. -1.786245 .4358719 -4.10 0.000 -2.644082 -.9284086

LIR -4.350146 3.009163 -1.45 0.149 -10.27246 1.572167

Covid 38.90095 55.50375 0.70 0.484 -70.33562 148.1375
_cons 124.0394 27.01418 4.59 0.000 70.87296 177.2058
 ABC Monday February 18 00:15:00 2019 Page 6

16 . vif

Variable VIF 1/VIF

LIR 3.47 0.288554

RIR 3.46 0.289431
NetFDI
D1. 1.05 0.951801
Reserves 1.04 0.961469
CA
D1. 1.03 0.974251
Covid 1.01 0.988051

Mean VIF 1.84

17 . xtreg d.GDP d.NetFDI RIR Reserves d.CA LIR Covid, fe

Fixed-effects (within) regression Number of obs = 300

Group variable: COUNTRY Number of groups = 15

R-sq: Obs per group:

within = 0.2110 min = 20
between = 0.0248 avg = 20.0
overall = 0.1123 max = 20

F(6,279) = 12.44
corr(u_i, Xb) = -0.0261 Prob > F = 0.0000

D.GDP Coef. Std. Err. t P>|t| [95% Conf. Interval]

NetFDI
D1. -1.154064 .2577993 -4.48 0.000 -1.661542 -.6465851

RIR -4.05018 3.289052 -1.23 0.219 -10.52469 2.424329

Reserves .6181038 .1892603 3.27 0.001 .2455442 .9906633

CA
D1. -1.554142 .3242965 -4.79 0.000 -2.192521 -.9157638

LIR 8.110147 4.436446 1.83 0.069 -.6230115 16.84331

Covid 79.35014 42.39383 1.87 0.062 -4.102244 162.8025
_cons 37.35639 45.70347 0.82 0.414 -52.61104 127.3238

sigma_u 244.27587
sigma_e 210.0525
rho .57490238 (fraction of variance due to u_i)

F test that all u_i=0: F(14, 279) = 19.05 Prob > F = 0.0000

18 . est sto pool

19 . est sto fe

20 . xtreg d.GDP d.NetFDI RIR Reserves d.CA LIR Covid, re

Random-effects GLS regression Number of obs = 300

Group variable: COUNTRY Number of groups = 15

R-sq: Obs per group:

within = 0.1590 min = 20
between = 0.6567 avg = 20.0
overall = 0.2631 max = 20

Wald chi2(6) = 104.60

corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
 ABC Monday February 18 00:15:00 2019 Page 7

D.GDP Coef. Std. Err. z P>|z| [95% Conf. Interval]

NetFDI
D1. -.9210913 .3511653 -2.62 0.009 -1.609363 -.23282

RIR 1.267604 3.473379 0.36 0.715 -5.540094 8.075302

Reserves 1.772065 .2213975 8.00 0.000 1.338134 2.205996

CA
D1. -1.786245 .4358719 -4.10 0.000 -2.640538 -.931952

LIR -4.350146 3.009163 -1.45 0.148 -10.248 1.547704

Covid 38.90095 55.50375 0.70 0.483 -69.8844 147.6863
_cons 124.0394 27.01418 4.59 0.000 71.09257 176.9862

sigma_u 0
sigma_e 210.0525
rho 0 (fraction of variance due to u_i)

21 . est sto re

22 . xttest0

Breusch and Pagan Lagrangian multiplier test for random effects

D.GDP[COUNTRY,t] = Xb + u[COUNTRY] + e[COUNTRY,t]

Estimated results:
Var sd = sqrt(Var)

not sorted
r(5);

23 . xtreg d.GDP d.NetFDI RIR Reserves d.CA LIR Covid, re

Random-effects GLS regression Number of obs = 300

Group variable: COUNTRY Number of groups = 15

R-sq: Obs per group:

within = 0.1590 min = 20
between = 0.6567 avg = 20.0
overall = 0.2631 max = 20

Wald chi2(6) = 104.60

corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000

D.GDP Coef. Std. Err. z P>|z| [95% Conf. Interval]

NetFDI
D1. -.9210913 .3511653 -2.62 0.009 -1.609363 -.23282

RIR 1.267604 3.473379 0.36 0.715 -5.540094 8.075302

Reserves 1.772065 .2213975 8.00 0.000 1.338134 2.205996

CA
D1. -1.786245 .4358719 -4.10 0.000 -2.640538 -.931952

LIR -4.350146 3.009163 -1.45 0.148 -10.248 1.547704

Covid 38.90095 55.50375 0.70 0.483 -69.8844 147.6863
_cons 124.0394 27.01418 4.59 0.000 71.09257 176.9862

sigma_u 0
sigma_e 210.0525
rho 0 (fraction of variance due to u_i)
 ABC Monday February 18 00:15:00 2019 Page 8

24 . xttest0

Breusch and Pagan Lagrangian multiplier test for random effects

D.GDP[COUNTRY,t] = Xb + u[COUNTRY] + e[COUNTRY,t]

Estimated results:
Var sd = sqrt(Var)

not sorted
r(5);

25 . xttest3
last estimates not xtreg, fe
r(301);

26 . xtreg d.GDP d.NetFDI RIR Reserves d.CA LIR Covid, fe

Fixed-effects (within) regression Number of obs = 300

Group variable: COUNTRY Number of groups = 15

R-sq: Obs per group:

within = 0.2110 min = 20
between = 0.0248 avg = 20.0
overall = 0.1123 max = 20

F(6,279) = 12.44
corr(u_i, Xb) = -0.0261 Prob > F = 0.0000

D.GDP Coef. Std. Err. t P>|t| [95% Conf. Interval]

NetFDI
D1. -1.154064 .2577993 -4.48 0.000 -1.661542 -.6465851

RIR -4.05018 3.289052 -1.23 0.219 -10.52469 2.424329

Reserves .6181038 .1892603 3.27 0.001 .2455442 .9906633

CA
D1. -1.554142 .3242965 -4.79 0.000 -2.192521 -.9157638

LIR 8.110147 4.436446 1.83 0.069 -.6230115 16.84331

Covid 79.35014 42.39383 1.87 0.062 -4.102244 162.8025
_cons 37.35639 45.70347 0.82 0.414 -52.61104 127.3238

sigma_u 244.27587
sigma_e 210.0525
rho .57490238 (fraction of variance due to u_i)

F test that all u_i=0: F(14, 279) = 19.05 Prob > F = 0.0000

27 . xttest3

Modified Wald test for groupwise heteroskedasticity

in fixed effect regression model

H0: sigma(i)^2 = sigma^2 for all i

chi2 (15) = 1.0e+05

Prob>chi2 = 0.0000
 ABC Monday February 18 00:15:01 2019 Page 9

28 . xtserial GDP NetFDI RIR Reserves CA LIR Covid

Wooldridge test for autocorrelation in panel data

H0: no first order autocorrelation
F( 1, 14) = 92.129
Prob > F = 0.0000

29 . xtgls d.GDP d.NetFDI RIR Reserves d.CA LIR Covid, panels(h) corr(ar1)

Cross-sectional time-series FGLS regression

Coefficients: generalized least squares

Panels: heteroskedastic
Correlation: common AR(1) coefficient for all panels (0.6970)

Estimated covariances = 15 Number of obs = 300

Estimated autocorrelations = 1 Number of groups = 15
Estimated coefficients = 7 Time periods = 20
Wald chi2(6) = 49.05
Prob > chi2 = 0.0000

D.GDP Coef. Std. Err. z P>|z| [95% Conf. Interval]

NetFDI
D1. -.2723173 .2589641 -1.05 0.293 -.7798775 .235243

RIR -2.956941 .8263647 -3.58 0.000 -4.576586 -1.337296

Reserves 1.029092 .2783027 3.70 0.000 .4836288 1.574555

CA
D1. -1.328304 .3240178 -4.10 0.000 -1.963368 -.6932413

LIR 1.231571 1.700285 0.72 0.469 -2.100926 4.564069

Covid -35.34976 15.71647 -2.25 0.024 -66.15347 -4.54604
_cons 22.15207 19.51389 1.14 0.256 -16.09446 60.3986

30 . hausman fe re,sigmamore

Note: the rank of the differenced variance matrix (5) does not equal the number of coefficients being tested (6);
be sure this is what you expect, or there may be problems computing the test. Examine the output of your
estimators for anything unexpected and possibly consider scaling your variables so that the coefficients
are on a similar scale.

Coefficients
(b) (B) (b-B) sqrt(diag(V_b-V_B))
fe re Difference S.E.

NetFDI
D1. -1.154064 -.9210913 -.2329725 .021813
RIR -4.05018 1.267604 -5.317784 2.843514
Reserves .6181038 1.772065 -1.153961 .1330504
CA
D1. -1.554142 -1.786245 .2321028 .0768615
LIR 8.110147 -4.350146 12.46029 5.254123
Covid 79.35014 38.90095 40.44919 16.33903

b = consistent under Ho and Ha; obtained from xtreg

B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test: Ho: difference in coefficients not systematic

chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)
= 138.52
Prob>chi2 = 0.0000
 ABC Monday February 18 00:15:01 2019 Page 10

31 .