19
1 2 Vol.1, No.2 2021 4 China Journal of Econometrics Apr., 2021 doi: 10.12012/CJoE2020-0001 (1. , 100190; 2. , 100190) , . , , , ; , , . ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; Understanding Modern Econometrics HONG Yongmiao (1. Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China; 2. School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190, China) Abstract This paper aims to introduce the philosophy, theories, fundamental content systems, models, methods and tools of modern econometrics based on its development history. We first review the classical assumptions of the linear regression model and discuss the historical development of modern econometrics by various relaxations of the classical assumptions to further illustrate the modern theoretical system and fun- damental contents. We also discuss the challenges and opportunities for econometrics in the Big Data era and point out some important directions for the future development of econometrics. Keywords non-experimental; linear regression model; non-linear model; model speci- : 2020-03-22 : (71988101) Supported by “Econometric Modelling and Economic Policy Studies” Fundamental Scientific Center Project of National Natural Science Foundation of China (NSFC) (71988101) : , , : , E-mail: [email protected].

Understanding Modern Econometrics

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

� 1�� 2� ������ Vol.1, No.2

2021� 4� China Journal of Econometrics Apr., 2021

doi: 10.12012/CJoE2020-0001

���������

���(1. �������������, 100190; 2. �������� �����, 100190)

� � ���� �����������, ��� ����������� ���!����"��#$�%&. ����'(�)�������)��'*"����+�, ,-. /!0"#+�, ��12� ��������$34, 516% �������� �����!�; &'( �)7*8'�, 7*8+�����92�,-�:., /;�����<=���0>1�#2�?3.

�� @45�;��'*"�;@��"�;"��A; 6ÆBC; 7D8#E; !9�; %&F�; G:HI�; J;�; KLFM; "�NOA�; 7*8; P<*8; :Q�=; >R; ?ST�; @UVI

Understanding Modern Econometrics

HONG Yongmiao

(1. Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China;

2. School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190, China)

Abstract This paper aims to introduce the philosophy, theories, fundamental contentsystems, models, methods and tools of modern econometrics based on its developmenthistory. We first review the classical assumptions of the linear regression model anddiscuss the historical development of modern econometrics by various relaxations ofthe classical assumptions to further illustrate the modern theoretical system and fun-damental contents. We also discuss the challenges and opportunities for econometricsin the Big Data era and point out some important directions for the future developmentof econometrics.

Keywords non-experimental; linear regression model; non-linear model; model speci-

WÆX: 2020-03-22

��YZ: �� ����� “���� ����” �������� (71988101)

Supported by “Econometric Modelling and Economic Policy Studies” Fundamental Scientific Center Project of

National Natural Science Foundation of China (NSFC) (71988101)

[\��: ���, ��� ��������������������������������������������� ���������, ���: � ��������������, E-mail: [email protected].

� 2� ���: ����� � 267

fication; normal distribution; conditional heteroskedasticity; endogeneity; instrumentalvariable; generalized method of moments; stationarity; structure changes; model uncer-tainty; Big Data; high-dimensional data; machine learning; forecast; causal inference;program evaluation

1 ]^���� �������, ���������������

, � �������. �, ��������������������

������� . ��������� �����������������, ��������������� �����.������������������� �,�����������, �����������. ����� ��������

���, ������ ���� �����������������. ��, ��

������������������. !����!�������"� , �

���������������������������", " ������. ������������������ #�, ��!�$# (2007)�"$���#� (2010).

�������!���%�$����%�. ��, �������%&�$ 40 �,

����������%'�������������&�, �������'�(��&�

�����. �!��Æ, ��������������, ��� ��������

�����, ������()�!"��, ����������� �� !, !��&����*") ��*++.

#, $��������#%Æ!���,�, �&&&'�"����#��������-����, $'. $��� �%,, $'./&��'�$��-'���������-���-'�Æ�(, $'.������-���Æ�0(, �%�����

����).�����/). �", !������� *&�1()�����, $2*�

����$����, ++.

,+�����, �� 40 �0 ������%��-1���, ., �

�����'�-��.3���!", Æ2'*������-�$��� �%,, �

�����-���'�����, !/0����������%�#(4�$1.

2 +_,-`./ab0

2.1 123c14d��!����, �����%�$����%�, ) �����Æ/���

������2, Æ$ 40 ��%%0�. ��, ��/ ������'�-����!",

�0���������!*1��. �0������!*+523�06�,-�-,

�,-3:

268 � �3 � 1�

1) 6�,-�-, . Yt = Xt′β0 + εt, t = 1, 2, · · · , n, � Yt ���, Xt ���/)�

� �6��/#0�� K- 4 �� (regressors) 4�, β0 K- 455��4�, εt $

�� �61&�7, 45� �� Xt �"#� 7�8� Yt �68�, ) n ,"�.

2) 9"Æ�, . E(εt|X) = 0, � X = (X1, X2, · · · , Xn)′ n×K 47. �*%,.9861&�7 εt � Yt ��58�$: X �*+. !*6�%,61&�7:' {εt} � X 2

;76.

3) %,((�( ;�, . E(εε′|X) = σ2I, � ε = (ε1, ε2, · · · , εn)′ n- 44�, ) I

n × n 8�47. �*%,.98 εt �%,($9�$: X �*+. �", {εt} :'$%Æ ;�. 7;�, !*6�%, {εt} � X 2;76�.

4) 61&�73:76(!<�;, . ε|X ∼ N(0, σ2I). %,!<�;-3<48 ε � X

2;76�. !<�;-3��# , (. ,"� n ��#9�) %,=����) 5�

�<8, �� =�����6��, �>7> (OLS) 6��, 61&�7 {εt} �9)�5, ? �;?�!<�;, � , �# ,%,=��0���)'�<�$�.

5) �� ,47 X ′X ��*047, �*%,� ,���!*:=, .+8!* ��$, 7 ���6�;9, �-5�:6��%Æ.

Æ X ′X ��*047%,=, OLS6��%Æ.Æ9"Æ�%,= OLS6��55��β0 ��.(6�.Æ%,((�( ;�%,=, OLS6����:6��.6�� (BLUE).

)@61&�73:76(!<�;�, �+.�# ,"� n > K, OLS 6���? �;

�!<�;, �*�# ,? �;���0;55��<@�Æ6��, &���0;��-

'�<����, Æ2=/��Æ t- <�� F - <����, ++. �!�$# (2011, A 3 >).

�0�����'/ ������!B?C. �0����� �������

��1��, �����!��A06�,-�-��0-3);6%0������'��, -3�>$�� �, �!D�, ;<��=�������-, �%��=������'��, �)��=%� ���?�>Æ, 9*'�-�&��9?, �9���.

����������� ���������)��, ��� ����EB��%. �='@, @���@*=A�0,-�-�'*-3, 0AB ������%��-1���!".

2.2 AefBCDEefBCFg�06�,-�-�61&�7!<�;-3, ����� �# ,%,= OLS 6�

�� ;�����? �;,:)<���).!Æ!<�;-3=,�0 t-<�� F -<

����Æ�# ,%,=F�3:�Æ t- �;� F - �;.

#, �=���B1���!*0-�G��C��!<�;, ��>D��, ���"�=���B1���?G5�� 3. � , ;6Æ61 ,!<�;-3��2��# ,�0? '�$H$�.�����C�� 1�$�<�6�,-�-H(@3:!<�;�<�, �<2)�06�,-�-'�@$�, �!*=/�H(!<�<�Jarque and Bera (1980) <�. ������!*���%�=061&�7�!<�;

� 2� ���: ����� � 269

-3, ��D$'�� (�������5�#�'), �I1� , (. n → ∞) %,

= OLS 6�� 76��3�55���!I6�, !��I1��6���D$!<�;. :� ,�����E , .�61&�7$3:!<�;, Æ76(�;�61 ,%,=, J�61&�7%Æ%,((, K@ ,"� ��, �0 t- <�� F - <�#$��,

) OLS 6��& BLUE. /C�, Æ%Æ%,((�%,=, �0 OLS '�Æ ,"�LA��, & $��. �*.�, @61&�7M(����4L%,((�, ��E�Æ

:'�61 ,&$��. !+D�$# (2011, A 4 >�A 5 >)

������ ,��5, Halbert White �������, 7 1984�12�2001�H2� Asymptotic Theory for Econometricians !�&�������� ,����0�F�.

� ,�;�����G8<, #�# ,�> ,%,=, ��6���<

�����D$�;�N��55��# ,�;�,;(FO, �Æ���)�HJÆ ��Type I � Type II ,(, II�).�$�I. ��B)D$�;Æ�# ,%,=�$;�G, �����C����C ( Klein and Spady (1993), Phillips (1977a, 1997b, 1977c), Ullah

(1990)) P�I+��%�# ,�> ,%,=�D$'�, �Æ2#G� Edgeworth

%J� Saddle Point $;. #, ��;@��, �&�6�� !"���. $�0, 6

8�H1C��KI)K, ������!*���% Bootstrap ��% !"��.

Bootstrap�'���& Edgeworth%J, #J78�8��H1�� ��)�=L��? )J�61 ,, ��B)����6���<�����N��# ,�;�$;�G,������ 5���I����).�, ��� �!"���. �=0���F Hall

(1992) � Horowitz (2001).

2.3 AhQiRSjklm�DhQnRSjlm��06�,-�-�9!*��-361&�73:%,(( ( %,($6 �

�:T���)��) �( ;�, .61&�7 εt �%,7L4$6 X :T���)��.

�-3%,=, OLS 6�� BLUE. @%,(($( ;�$�6�, $M OLS 6��

$H BLUE, �0 t- <�� F - <����&$H�3:�Æ t- �;� F - �;, FUÆ� ,%,=�0 t- <�� F - <�&$H$�. �0����� V��� %,((�( ;��N*-3� #�, �� 1�!1�>7> (GLS) '�. GLS '�-361&

�7%Æ%,0(� ;�, #%,0(� ;��EK%5� (J%Æ!*55��),

�������/%5EK�%,(0F5%,0(� ;�,H/&�6�,-�-W��!*4L%,((�( ;��6�,-�-, :)��/D�6�,-�-)� OLS 6

�, ���06�,-'���$�. �, ÆO<�Æ:'6�,-�-�, �61&�73

:!*E�L�� ,-��, K��� Cochrane-Orcutt F561&�7� ;�, �

U�D$�G6��.6�� GLS 6��.

#, “%,0(� ;��EK%5” �*-3)�$$9�=���� ��. Æ�&���, %,0(� ;��EK55�. Æ( ;�%,= (��FP5��), �Æ

�����!I�6� OLS H(�%,0(, H; GLS 6��GK�, :)� $�

270 � �3 � 1�

��� GLS 6��, �*� ,%,=�� BLUE �X (�! Robinson (1988), White

and Stinchcombe (1991)).

#, �&���, �M���#.H*�NQG8� OLS 6� ;�����)

. White (1980) �I�%,0(%,= OLS 6���D$(GK! 1� 6�,

�OP� White’s heteroskedasticity-consistent variance-covariance matrix estimator. ���*!��(6��, �H!�0 t- <��, ��Æ%Æ%,0(# ,"�LA��, I��

�$�. #� �<���P�ER� (robust) t- <�. 9!5, �0 F - <�����H

!, ��Æ%Æ%,0(�, $H$�, .�� ,�&�. #, �0Y��!�(6���ER� Wald <��SIT<>$<�.

��Æ:'6�,-�-, @61&�7$#%Æ55EK�%,0()�#%Æ55EK� ;��, OLS6���D$(6��$M�F=%,0(�*+, &�F=55EK� ;��*+. �J�6�#G��>( - K(47. Newey and West (1987, 1994) �

Andrews (1991) + 1�����+6��>( - K(47, ��!"��������. #, ������H1�?��� , @%ÆQ@� ;��, ��+6���>

( - K(6��, ��II;����<��Æ�# ,%,=&�GUV!����/-3, .%Æ �� Type I ,(, �*� UW6�� QR�/X, L�%�'*B#.

2.4 AopqrDEopqrÆ������, 6�,-�- (Yt = X ′

tβ0 + εt) Z��� Yt � �� Xt 55��

β0 �Æ�6���, � �� Xt �A*$A���/)�� �6��/ (�7$

��) #0�, ��$Z����/&/)���Æ�6���. �, @����%,5T($,-J�) A*��/)���!*=7K�, �3OM�6�,-�-. #, Æ$�������-�, @����%,5T$55���6�J��, ���� ���Æ���!D�6��. )�, @�-$YK����%,5T, )YK����%,(, %,�

��, %,4FU9*%,�;�, �!*�6���, $L���$[/&��/)��)C, S�.

Æ�Æ:'������, 6��Æ:'�-��Z����/)$L �� 61&�7�Æ�6���, �#G� ARMA �-. ��NQB���6��Æ:'�-Æ2 ,-�Z�- (TAR), C�M[1\WM�- (MCRS), �EW/ ,-�- (STAR) +, ���-5��Æ:'����%,5T)��6�;�. ���6�5T�-�/)��$(]<=�6��-N9;��, ÆA!*]<=, ���!*6��Æ:'��, Æ9!*]<=,

���9!*6���, )�6��G�����$(]<82;W/�1\-3)�. �

6�5T�-���YK��B1�����6���, ���^>���P��B1NT�Æ(����P�.

Æ 20 O\ 70 �, PNQ1, O�PÆ\G� D(UQ8Æ(V, IIO*��64$���. 8 G��$��� ���B1NT�*+<��@�!*EF���� . �

�9���������, �����C< 1'*7-�%,(�-, Æ2 Engle (1982) �

� 2� ���: ����� � 271

ARCH �-, Bollerslev (1986) � GARCH �-, Nelson (1991) � EGARCH, � Glosten et

al. (1993)��Z GARCH�-, ++.���-!$�9*%,RÆ�;;�, J�J�A*�Æ:'�����7L%,4;�, ��$,���;� (MLE) 6�55�-��. Æ�����, ��6�9��-�55��, !DJ�-3W"�S_%,, :)�I1�Æ:

'����9*%,RÆ�;, ���� MLE 6��-��. ��S_-3�,$!�(��[CÆ&5X�!�), 9��-�;�J����,,3, �*OP� Quasi-MLE,

$ QMLE. J��7L%,4�-3�!�, QMLE 3�!I�6��-��, #] 6�

��D$(H����!�%,R�;� MLE �D$(, �� QMLE 6��NQ$�

�. QMLE D$(�.0, � MLE �D$(.0�)=�$(, J7;�6�,-�-�OLS 6��Æ%Æ%,0(� ;���D$(.0, ��YO��;�J�,3�3�

$��ER(GK !I6��. ) MLE �D$(, .0K7;�6�,-�- OLS

6��Æ%Æ%,((�( ;���D$(. �6�,-�-� F - <�7;, @;�J

�,3�, =/�;�N<��$H$�, ��J;@���� MLE �D$(GK. ���

�0;�� QMLE D$(GK�ER��<��, ER Wald <��ERSIT<�$<�. �!�$# (2011, A 9 >).

9!7������- (Æ2 =6���6��-) �!*$!;6-4%,0YK. 6-4%,!D:��'��1, �, T����'�L>'�.9861`J�],(;�����-��@P�%,5T�(, .G*�>5$%Æ���].(. U��*�X, �

Qa!�9$�����, 0Y!;6-4, @ ÆN���TZ:T�, 6-4�(. ��$5X�����9-%,�;, ��� MLE. !14 (GMM) 6���,.3, �0;!

; ,4, � ,4�4G$>�55���4G, �DQaH ,4^�_$6-4���T, ��55��T�6��. ��2, GMM 6��� ,4�!*9)7L7�>�, ��)�!D&*+ GMM 6�����G. !*D$�G)� ,4�( - K(6��,

��� GLS 7;, �F5 ,4�0(� ,4��;��, :)V�D$�GGMM 6�. �*� Hansen (1982) 1. GMM �M������046��R%,

#J��6��<�����-, �'�L>�-) 1�. GMM ����!", �=

������6��5�M� ��, Æ2 OLS �7L��>7> (2SLS). QMLE � GMM

6��6�������-�N*���.

2.5 AstpDutp�06�,-�--361&�7�� ��Æ'*�>$L:SIT� %,5T�(,

. ��Æ@>��J$H0�+8:T5$*+61&�7��58�, �*%,P�9"Æ�. Æ������, �=*��"Æ���1. �61&�7:' {εt} � ��:' {Xt}2;76, $ ����61��, KP�%Æ@"Æ�. �61&�7 εt ;��@> ��Xt, $L:SIT, %,5T�(, KP�K"Æ�.K"Æ�.986�,-�-%,5TE(Yt|Xt) �!�3�, .%,5T E(Yt|Xt) = X ′

tβ0 �� Xt �6�J�.

�06�,-�-�9"Æ�-3A�@"Æ��K"Æ�N*%,�Æ, @"Æ�.98

272 � �3 � 1�

9"��6, )9"�.98K"��6, #W�$�. �06�,-�-�#�-39"�%,, ��<�I�# ,%,= OLS 6�� ;�<�����? �;. ��

,D$'�, K"�%,�LA�.

@K"Æ�$�6�, 61&�7;��@> ���%,5T$�(, �*LE��OP�%Æ!Æ� (endogeneity), �� �� Xt P�!Æ��. JÆ!Æ��/� =, Æ2 ��%Æ �,(, %ÆRM��, %Æ(6�.(, ++. #G(6�.(Z5�#F=�6�,-�", #�!*$X*RM�, �*$��RM�N, ��8���� 7��X�. Æ�*L`=, ������!D��T4����, . ��*+���, (�

���&*+ ��. �����C�H�*T4����II� ���61&�7�Æ%Æ�;��, P�!Æ�, ���.98 �� Xt &!*!Æ��, �(9�;:(X�.

9I�', !Æ�II!�%, E(εt|Xt) = 0 $�6, :) OLS 6��$55�� β0 �!

I6��. �YZ1, 7/�, ��%Æ �,(, ,-�-%ÆRM��, $J�EK,

3+, &&II E(εt|Xt) = 0 $�6, #9I2'��/��!Æ���. G8%!, !D�P

E(εt|Xt) �= 0 �%Æ!Æ�.

@%Æ!Æ��, �J� ��X�����6�,-�-)�6�, OLS6��H$

N���T�!I6�. ��, �*� 2SLS 6�, a_!;�61&�7$;�#� ��OF;������, ÆH �� “*U” ����, �DH���� “*U��” )�,-,

� �V�84���� (: �� ���) ���T�!I6�. 2SLS6�%�$����

, �V 20 O\ 20 �Æ['�������%%0�, �b�P�Æ�8Sb�8����� (�! Stock and Trebbi (2003)).

�����, ��� #�������� ��;�� \, II 2SLS 6��$E�

FU$,!I6�55N���T. �*L`P� “K����”(�! Staiger and Stock (1997)).

K��������� 20 �0������!*����4.

!Æ�� $MÆ6�,-�-%Æ, Æ 77-��- (%,(�-�%,����

-�%,�;�-+) �&S�,%Æ. 9", !Æ�����]��,-�-&� ��'(�! Blundell and Powell (2004)). !Æ�� Æ������c�+5��, ��/�, Q

���C����+T��6�����������, :)��������. ��

', ����, JJ�;��, $J�����. �!�$$9����.

���� ����� ���������, ��$,���\ 7��:T$�, 8�

��V���A*$A��������@(%������. ��, 8������

������!*��P�, & ��������!*R�� . W������ 20 �

0�!*��R�#G� “Z'8� (treatment effect)”, �*SU���, ac�����

ÆV/W����61X� (randomized experiments)�.3�,�%1�!Y��6�������������'��, ^Æ�(VT6����� (econometrics of program

evaluation) �*BX��, ���Æ����%,=��6�Y?C�0��T6'*��*&G:(V. ��T6!*��(V�8�, �,.3Æ(+%,=, NQ�ZY(V�.��-36��Z�7(V�Y?.�, N[�(�Y7(V�8�. �d��b�P�

� 2� ���: ����� � 273

Æ(V%�Z�%,=, 8d�6�-3(V6��Z��Y?.�. $�, Æ2T(�

(difference-in-difference, DID), ,-)�3� (regression discontinuity design, RDD), U4T�

VW (propensity score matching, PSM), 5_��T6 (panel data approach to program

evaluation) +, %!"���'*(V�7�T6 (�! Imbens and Wooldridge (2009) �e,+>). �, Hsiao et al. (2011) 1!*��5_���(VT6, !���T6ZZ�!

� 2002 �X�� “��;6�eO�Y���`- (CEPA)” �ZZ��f��*+. 9!

5, $ 30 �X%������, ���\��[%,�����-���5'��, &� �

������!*B�B��. �", #� “\[�� (field study)”, �!*d��,

Æ �*&��[f=,��(;!����\V��, \g�������Æ�����.

2.6 AqrevghDqrwg@6�,-�-%ÆRM��, $J�EK,3�, @�P6�,-�-%,5T�,3.

�-�', �%Æ!*55��T β0, ��6�,-�-�%,5T;+, . E(Yt|Xt) =

X ′tβ

0, KP6�,-�-%,5T�!�3�, )�*55��TP�N���T. �$

%Æ+8!*��T, ��6�,-�-+�%,5T, .�#���T β, � E(Yt|Xt) �= X ′tβ,

KP6�,-�-%,5T�,3.

@6�,-�-,3�, �-��$,O/)�N���T, �$,O]\��;1. �,

6�FÆJ�#%Æ,3, K��T$,/)�a&FÆU4, ��a&FÆU4FÆ�];

�.I�, @FÆJ�$6��, a&FÆU4$+�!*��. 6�'0, $L�%,5

T$%,R�;� 75 (%,(�%,����QU9*%,�;) ;�, #%�-,3, ��<$,/)�N��-��T, �$,O]\��/). ��/)��8�:X�

�-3��!��@. 5���/)��8��", �-,3Æ�&���&�,&II9�D�. �, ÆB1NT, ��,3�-&II#G� “�-^] (model risk)”. $��H 2008 �

^gZ[�D�LbQ1-h�_`i@�\c�����B1_ÆJ]�]�Q^j.�-

(Gaussian copula model), Y�-$,!�YKB1NT����P(���.

#, �-,3!$.98�*�-!�Z$$,�&���. �, -3A*���

���<`���: 80% �6��8��: 20% ��6��8:(X�. )�, 6�,-�-

d,�, #J3�$d�L ,+, L� ��$,/)� ���a&8�. 9", ��,3

�-�M���3�]\��;1. �, Æ��%',�Æ�6�,-�-�, ����`$];, � ��JÆ2%'�G�����, �*�-�,%ÆRM��, ��� $ �

*�,+&&*+�`$];, )*�,+�%'�����;�. #, ���[�X^J%',�Æ, �*�����H%'�G�*/)�����T!I6�10, � , R

M*�,+�!�����6�,-�-& T���.

�-,3#�,1 Æ�=LE. !*6�,-�-, �� ��%Æ#(�a, %Æ__�� (censored data) $P)�� (truncated data), �I OLS 6��<$N���T�!I6�. Æ�*L`=, �f9S_-3, �-361&�73:!<�;, �D� MLE

6�55�-��. ����%�a, ���� ������C�2%���6��

274 � �3 � 1�

�, # MLE ��H/&6�,-�-���!I6�10. �d, S_-3@!�, &

*+6�,-�-��6��!I�. 9!5, 9��-������7L%,4;�. Æ9��-!�3��L`=, @�!$5X����%,RÆ�;, ����&J�f9S_-3 (-3�d�61&�73:�d!<�;), �<�I1����%,RÆ�;, :)��

MLE 6�55�-��. ��S_-3�,d,�, %,�;���,,3, �*P� QMLE. #, �%,5T�%,(3�!�3�, K.�%,�; (.%,QL4) ,

3, QMLE3�!I6�%,5T�%,(�-��55��T, ��2,6�,-�-����a)J�S_-3�LE$(. #, �-,3H*+ QMLE6���($��G, ��

J���ÆRÆ�;�-,3�&$��ER�D$(GK. �! Bollerslev and Wooldridge

(1992), Lee and Hansen (1994), Lumsdaine (1996).

��,-�-,3 (. E(εt|Xt) �= 0) *+ �����/)��8�� ��6��!I�, Hausman (1978) 1�!*<�,-�-,3�,OP� Hausman<�. White (1981),

Newey (1985), Tauchen (1985) � White (1990) H�R% �4%,�����!D��-3�<�, P�4<�$ m- <�. ��<�5$!I<� (consistent tests), .�<

1!�,3�-, FU� ,�&�. Bierens (1982, 1990), Hong and White (1995), Fan

and Li (1996), Hong and Lee (2013) +*����+ 1�6�$�6�,-�-3��

<�. @ ,"�LA��, ����< +8EK��-,3, ��!I<�.

Æ���������, �-<�#�9!*EK, P��-�� (model validation). �*H����bk�� (training data set)� X�� (test data set), �bk����6��-��, X��K��T6%6���-�9!*���L ,+. �*�-�������-� ,"L ,+, Æ�Æ:'L �1`�c�!"��. ,"�-����ea ,!�-�G?9 (overfitting). �", �JÆbk���Æ����JÆ X���

����)=$( (��.0��$ ,�-(0�#I), �I.<�-�bk����

��3�!��, J�����L �,$&`H.

2.7 AxypDExypÆ-3c��E�Æ:'61 ,�%,=, �06�,-� OLS '�, @61&�74

LM(��%,((�, Æ� ,LE=, #$��. �%Æ%,0($ ;��, OLS

6��.�� ,%,=&$ BLUE, �0� t- <�� F - <�&$H$�. #, ��H

!�0��<��, � t- <�, H!D�ER��<��� ,����.

#, 2, OLS '�Æ���� ���Æ:'���E���, !D$H$�. Granger

and Newbold (1974) ���H1�?��� #Ga,- (spurious regression) �, .N*

2;76���E8�U�Æ:', �H �!*���9!*��)�,-, K,-��6��� t- <��Æ���2)=�. Phillips (1986) :'�2a�9?�/). Nelson and

Plosser (1982) ���� Dicky-Fuller <�, � �M�T���B1�Æ:'5��E8�

U��. ��, �����C 20 O\ 80 �%<I+��%��E�Æ:'�����'��, Æ2 Engle and Granger (1987) �K9'�, Phillips (1987a, 1987b) �8�U'�, +

� 2� ���: ����� � 275

+. �6�,-�-, 8�U�Æ:'� ,'���E�Æ:'�����'�bZ$(. d

eAB��F Hamilton (1994).

58�U���", ��E�Æ:'&�4 �$1�E��. #G$1�E:', Z 5T�Æ�J�, ��$!*�E�Æ:'. @5T�Æ�6�J��, �*�Æ:'<&)�!*�>6���$1. #H�Æ$1F5, �Æ:'H���E��. � ,"�$

�, $1�E���8�U��JÆ�����7;, $"f��. ÆT��������, 2

)T�����8�U��#$1�E��, �����,��J���$(�(V41.

�, ��� T�f(bg(V@���>8� (@;�����8�U���) $J�h>8� (@;������$1�E���).

��E�Æ:'#�4 �.0�� (structure changes). .0������N*, !*.0c� (structure breaks), 9!*cb.0�� (smooth structural changes). %Æ.0c���Æ:'#JÆ���, �8�U��JÆ�����E2J��7;. ��.0c�, Chow (1960) 1i�8*%5�� (change point) �.0c�<��0, �����CÆ�*5:���L��%, �jg=*��, � $5X���LE=8<�@%Æ.0c�, !�;6�NQbd�� ,�;'�. ��F Andrews and Ploberger (1994, 1995),

Bai and Perron (1998) ++D.

5�.0c��", 9!*.0��cb.0��, .�-��6�Æcb��. !*�

$#G���� STAR�-.$1�E��&!*�$,� 5T���dJ��e.

������l.H�C�)K�(V���-\Bf�"M%,��+/�, ��II�������.0���. �", �����-��'�L $g5(V+��, !$���B

� ��, �&&II��.0���, ��=/�cm^d2 (Lucas (1976)). #, .�

2,�8c��Æ, ��ch�i9�,+/�, ���-��� ��.0&�,&cb���. FU.���*-����Æ�c�, �*-��c���Æ�$(, K96�T���:'&&1 cb��<1. 6�, =LE=, cb.0���,�_$�� �. !*0-j

�D�T���9�� “�c� (Great Moderation)” �. D� 20 O\ 80 ��>�0, T���9�, � GDP f�Æ��bekÆ�9�, k 1!*@De>�$1, �P� “�c�” � (�! Bernanke (2004)). �, ��T���&%Æ7;� “�c�” �, ��� GDP f�Æ��bekÆ, �: 1992 �� 1999 �%1 �9�@D�>�$1 (�!

Sun, Hong and Wang (2019)). #, ���T���� “�c�” �, ;iUW#6��f;�� /�.

Chen and Hong (2012, 2016) 1�<��Æ:'6�,-�-� GARCH �-@%Æcb.0���, �!*����� , .0���,`J]hÆL $d�!*��/�. cb.0��&�,JÆa�ni � (spurious long memory), .��B1�Æ:'�

���ni ���,�cb.0��II, )�N!���ni�X. Hong, Wang and Wang

(2017) 1<�9�E��K�E��, !���<�T���B1�Æ:'��E�. 7

����4f, (�D�T���B1�Æ:'!$4L�E�-3, � 5T$(6�Æcb��. (�$J$1D��Æ:'4L�E�%,, ��E�Æ:'������!*

276 � �3 � 1�

�,-3.

Æ�Æ:'���, 20 �01 �!*B���SU, . M�E�Æ:'�� Dahlhaus

(1996). #G M�E�Æ:', !*��E��, ��T6�Æcb��, ��, Æ!>��Æ!, ���!*�E�Æ:'�-0$;, #$(�>J��$(��E�Æ:'�-0$

;, )��$(�-���T6�ÆjjcbB��. Qg���V��� Q, �_�D$�.0������ M�E�Æ:'�-���.

2.8 A3c14dqrD14lz2,AB�, ;iE$ ��'������;��#%���. @�E , ���

���'*'K��-, Æ2%,5T�-, %,(�-, %,����-, %,4�-, %,

RÆ�;�-, #��C�k�!16�,-�-, Probit �-, Logit �-, Cox N�Q]Æ

�-, lA,-�-+. �����, ��h!7�-J��#������� �,XX�,

. “� I4”. !$!*������-$�����+8��� . �, ��B1N

T�8�$�L ��, J��`J,��%,5T;�; ���-���B1NT��9�l18��, J��%,(;�; )��^]]T (value at risk, VaR) �, KJ��%,���$%,�;;�.

��'����#- Æ;��/)���Qa2. Qah�����/)��, 5���

�� ��������������m,�", #J�II��'�, ����'��m

_2)����������. �-3�!��, ��'�#���-��]\��;1,

f@�-��/)�. �������-,g� ����#� ���.�, :���Mo)

�/), U����. ������, 8i������������m,$A�,

J�a_��'�. �I, 8<���'�@,A/)#� ��� �? /C�, 8<

���'���8�$!��? !*�,.3H��'�$��-'W���������-�!*:=, �D����������<��*:=@!�. T�@i�, �*�����, !DJ�f9!�S_-3 (�3��-�J�EK). ��, @UV/-3�,

YO���p����'�$�6, #��S_-3$�6. !D'0, �=���'�$

��-'5� model-free, .$Iq�A*�-EK�������-, @@�H��'�$�

�-'W���#Qa�������-���:=�, ��-'���-'N[�Æ&%Æ(. /C�, N*-'!$+], ��Æ/)��-'�<�.��J���?h. �<�NT

�8���, �`J]h�@P�H0`J]h�%,5T6�+8L ,+, @�PK

K�8NT-'�6 (Malkiel and Fama (1970)). �*��-'!* model-free �-'. Æ-3Qa!* p L6� ,-�-<��*��-'. @KK�8�-'�6�, ,-�-6�L ,+, #� ,-��@�(. �� U��!* ,-��$�(, �IKK�8�

-'$�6. #, � “#� ,-���(” �*��-'6�OUV, @�,@'KK�8

�-'�6? $,. ���6� ,-�-JL `J]h�n=�-��!*)%. #

� ,-���(, J,'f6� ,-�-6�L ,+, $,'f`J]hÆ����50`J]hÆ6�L ,+. ��, �#� ,-���(, @�J,'6�� UVKK�

� 2� ���: ����� � 277

8�-'���. 6�, J�'.��-'���-'�(, !���������-��, N

[!$+].

���������'�EB��%5, ��8�b��. 68���$)jr, �

B�������$)JÆ, @�&NQ"fb UV ���'����, . ���'

�$,/)� ��, ��<J�B���'�. ��'���%�� !*��, �*���, �����,g&V���%.

������ ,"��#�#��������� #�+'*/�, ��k�,%

ÆN*$N*�2���'�$��-', (������<�, /C�, %Æ=*��'���/)(!�� ���,�. ��, Rh*��'�!��? �*LEÆ���P��-$��� (model uncertainty) $�-�l� (model ambiguity). Æ����, ��%Æn=jÆ�/)��$L ��, ��� ��,�1 :6� �. Æ�&���, �,%ÆX*$(�������- (Æ4$(/)���6�,-�-, $$(J�EK��-), U�!

�����dK, 4 ��;$$Xi! . �", ��,��A*��, !*������

-4 �H, #����Z9 “W&”, f9$B�X*���, K4 �:��-�,&��9!*. �* �, Æ���&P��-$���, 7;���'��$���. �=0��!

Breiman (2001). Æ%Æ�-$���%,=, 8/)�-$���� �� ����)�

*+, !*�����s . �$MM�� �-�/), &�� ������!��

,-3, ��-i!�-3, .%Æ!*i!�������-!�YK��Æ���. ÆL SU, �����-;9$�-�50t��-$����L �*+.

2.9 Ao{|pDq|p682(j�M�2(j+@PC���%�!"��, ��� ��!�Z�f��IG

KIf9, �#G���� (Big Data). 5�/��r-.0����", ���#Æ2��

��.0��].0���, +,�kE�mg�Mg+��. .��r-��, &Æ4!�

B-��, ��Æ���J����mn��+. ���0sÆ22(jnlGm��f���kn�(oM��jt, '**��n�g�`, ++. ������=��$Xi��]pnl, ���-��, # =L`=@POG\.

��� 5�/���#6�� =@P, ��@P���00;��$,�� G�!�����. �, ��*��n�Bom-+,��, �0;p`[$FÆ[�LoZ� (

Baker and Wurgler (2006, 2007)) ���(V$���Z� ( Baker et al. 2016), �D������������B1NT�*+.

���&��0;Qg������Z�$�������,. �, T���V]Z�,

CPI $ PPI, ��? �QgÆuG��. ��2(jFÆ]�ÆJ]]I@P, bZ��,

a_��m,0;G^FUGq�FÆ[V]Z��ÆJ[V]Z�. ��QgT��������T����@>$$>���]`, ����� �.1. ��Ln, Æ$O�H0,

QgT����H��,��)Æ, � �BX��,A�H���T����B1NT�Æ�2���. C�2, ���%oÆ���L (nowcasting) �����. “��L ” �*�ps

278 � �3 � 1�

�o��, ��Z 2 >��!�h>qoL . ������ “��L ”, Z8������L @q GDP JT���Z�. ��qG GDP �G;�Æ�Q��vD>, ��L

@q GDP +��T���Z�, �u�T���������.1. $��C��qr�%J&�' �C� GDP )���L . ����L ��=0���! Giannone et al. (2008).

(Wang et al. (2005)) 1�!* TEI@I L �, H+,rv���;����m,+e9p�, /X�H������0X����������;.9�P , )=� !� ,"L �G.

���J�$���/����B-��. �,Gp�f���<!`J]I�������J���+. ��B-��rsB�������-�. �J 20 �, Qg�qQg

B1��%����QgB1����KI�%. �, Engle and Russell (1998)��qQgG

p�f���]I���Æ��ÆÆt)�;�, 1!* autoregressive conditional duration

(ACD) �B�-, �*�-�8�qQgB1�f��L =!*]I����Æ, � 50

G*�Æ�]I���RÆ, ��L �����p`�fVp3� _Æ]�]. 9!5,

8�<!`J]hÆ��, B1�����C 1�B�6�G<9�Æ G<`J]hÆ�Æ�K(�, ��Qg9�Æ�K(��0Bdp`;9L'�B1^]L'. �!

Aıt-Sahalia et al. (2010), Andersen et al. (2001, 2003, 2004), Barndorff-Nielsen and Shephard

(2002, 2004), Noureldin, Shephard and Sheppard (2011), Shephard and Sheppard (2010).

���J�!7OP�J����B-��, GqsG��q6, 2NGmGqur]

Iq6, ++. 5_��J����!*��. �7�����2o�!;B�-, wJ�

���- (functional data models) (�! Muller (2005), Muller and Stadtmuller (2005), Ramsay

and Silverman (2002, 2005)). 9!7B-����� (interval-valued data), .�A*��

��QT��\T#;����p9. /����=���� (point-valued data). ;����

0', �Æ��Æ48�=@P, #�>�06�� �88�. �Æ��!$�!, �Gq�

�Qs�\s,Qst\st,Gqur�Q]�\],B1`J�st(] (bid-ask spread),+

+. �Æ��#Gmn�� (symbolic data) �!*��. Han, Hong and Wang (2017)ÆÆ

1������� autoregressive conditional interval (ACI)�-,8�����L H0

�Æ�>. �*�-�M��Æ:'��� ARMA �-��Æ2,. &�! Han et al. (2016)

� Sun et al. (2018).

���Æ =L`= ,"�;@�, �Qg��. ��� ,"�O������jÆ/)$L ���4�, �7��P� “Q��� (Tall Big Data)”. �/���;N, Q�

���_�rv������������, ��6���. ���6�������-

Æ�����t��!��%���>Æ. �������� =1`�c,�XVu,

61uv, ��v�jw, fG�c+, S�6��H1H, ��G27;�������

���. � White �4������C, 0;���v�jw���;��'�� (�!

White (1992)). ��v�jw�-PO!"�����B1��������. C�2, ���

�����G!$Æ� ,"��, )�Æ�u� =jÆ�/)$L ��, FU��,1 jÆ/)$L ���4�q� ,"�,�P� “x��� (Fat Big Data)”$Q4���. Q

� 2� ���: ����� � 279

4/)�� 5���Qa$�-Qa� =�,�. �Q4���, =LE=�,J���55�/)�������/)$L ,+. 1`�c%�f$��H1H (XVu�61

uv+), 0Qa��/)$L �� (�! Varian (2014) �AB���). 8��Qa��/)��!6� *+, J�a_1`�c����, !������'�.9%0. �5�!*0-j�:Q4���Qa����� LASSO , �.91`�c�!* “�

��c” (�! Hastie, Tibshirani and Wainwright (2015)). ��Q4���, �.91`�

c)���������'�� �, Æ2: 8Æ��jÆ�����Sb�)=�����, �r �:����6�8�, :)/XK����s0�x&; 8Æ����SÆZ�� ( *��,q� ,"�), ��Qa ����L ,+�SÆZ�, r �: ,"L 8�; 8Æ0;`,`J�]�-�, :��jÆ�^]�$�, ��G*`J���X*��^]�$!6� *+, � 8ÆQ4���J�6�� (�%Æ=*/)���), ��1`�cy\���J�6���4G, �� �H�6�8�, ++.

����1`�c&�t��������5 5 ��m_. �+#,, ���

������ �����(VT6�!*�,.3, Æ(+%,=, NQ�ZA!(VD�.��-36��ZY(V��Y?.�. ��(V%��Z, -3Y(V6��Z�Y?.��� �, ��J�������06�$L . )�, (V8�T6�d��:X�6�$L Y?.�������������8�, ,XÆA*.12�M� ,"L . ��1`�c���������� 5�v�@P, ���H1H 5NQd�� ,"L , ��L�, ������1`�c�L ���d(VT6 5B����.

1`�c�HL !$!�������������, C�2HL ��E2J�*

“wv”, $�t yu��/)�. #, ! Varian (2014) Z1, ����1`�c�Y?C

����L ������������d��T6(V8� 5z�m_.

�Y@i, ���!$4Z ,. �, �����������0X�����, ��$

���, ��hz6�8�@�#���, &$,��)� ,"L . ���� “x��

�”, .jÆ/)$L ���4��� ,"�, � �Q4���:�����;��oGE, � “> ,”. � “Q���”, . ,"�O�������/)$L ���4�, @

�Æ;�����)�, ��$Y�/������ , ���'�61? JÆ���6�$���.�,@ ,"��XwFUX�w�,-3��6�T� t-<��xHÆ 1%$ 5%)

=��2)=�, � �/)���p,A/)������=�N�? ��I� ,"�,

!*����)=��/)��@��8�2&)=�? � ,%,=, �-��6�

�<������,$ �-Qa����, �@%�-$����, �-$����*

+�,ON�-��6�$����*+����=. ��, Æ���;��, �Y��'�-Qa��-$���.6�,���$�= “�”,6�������!* ,@P,;6Æ? '���2��������;�, �,��.3!$��, #J�EB��%.

3 w},+:�0�������06�,-�-1�, ��=A6�,-�-��0-3,

280 � �3 � 1�

AB� 40 �0 �������%��'�-�, !0������-1=�����#(����%4�$1. ��E1, �������0������A0, " ��

�!<�;,%,((�( ;�, 9"�,�-!�3�, .0�E�+�0-3�A0, �

�R%�������$�����?, yz��������-����, !����

��'��9?�����, �������� ���������������, Æv)������, �)��'��EB�%���, ��8���b��. Æ����,

�����#Hw0����%.

,+U��[�"��Yf���&��Yf�x��qf������x�_wx�����L'�x�{J��B1�%��x�x���yUz�����5 2018 �����

x`|bx�wyf������x�x���yy{����x 2018Z��������

z>�z�2v*&��x��Æx 2018 Z���Æ�����z>�z�`{9')�.

�[g{�C ����B{{& “��;����(V��”�����57� (71988101)

�`_.

| } ~ �

���, (2007). � �z{|�~|{|} [J]. � ��, (5): 139–153.

Hong Y M, (2007). The Status, Roles and Limitations of Econometrics[J]. Economic Research Jour-

nal, (5): 139–153.

���, (2011). |}� � [M]. |}: |}�~~�}.

Hong Y M, (2011). Adcanced Econometrics[M]. Beijing: Higher Education Press.

~�}, ~�~, (2010). � ���z~��|} [J]. �� ��� ��, (9): 133–146.

Li Z N, Qi L S, (2010). The Strength and Disadvantage of Econometric Models[J]. The Journal of

Quantitative & Technical Economics, (9): 133–146.

Aıt-Sahalia Y, Fan J, Xiu D, (2010). High-frequency Covariance Estimates with Noisy and Asyn-

chronous Financial Data[J]. Journal of American Statistical Association, 105: 1504–1517.

Andersen T G, Bollerslev T, Diebold F X, Ebens H, (2001). The Distribution of Realized Stock Return

Volatility[J]. Journal of Financial Economics, 61: 43–76.

Andersen T G, Bollerslev T, Diebold F X, Labys P, (2001). The Distribution of Realized Exchange

Rate Volatility[J]. Journal of American Statistical Association, 96: 42–55.

Andersen T G, Bollerslev T, Diebold F X, Labys P. (2003). Modeling and Forecasting Realized Volatil-

ity[J]. Econometrica, 71: 579–625.

Andersen T G, Bollerslev T, Meddahi N, (2004). Correcting the Errors: Volatility Forecast Evaluation

Using High-frequency Data and Realized Volatilities[J]. Econometrica, 73: 279–296.

Andrews D W K, (1991). Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Esti-

mation[J]. Econometrica, 59: 817–858.

Andrews D W K, Ploberger W, (1994). Optimal Tests when a Nuisance Parameter is Present Only

under the Alternative[J]. Econometrica, 62: 1383–1414.

� 2� ���: ����� � 281

Andrews D W K, Ploberger W, (1995). Admissibility of the Likelihood Ratio Test when a Nuisance

Parameter is Present Only under the Alternative[J]. Annals of Statistics, 23: 1609–1629.

Bai J, Perron P, (1998). Estimating and Testing Linear Models with Multiple Structural Changes[J].

Econometrica, 66: 47–78.

Baker M, Wurgler J, (2006). Investor Sentiment and the Cross-section of Stock Returns[J]. Journal of

Finance, 61: 1645–1680.

Baker M, Wurgler J, (2007). Investor Sentiment in the Stock Market[J]. Journal of Economic Perspec-

tives, 21: 129–152.

Baker S R, Bloom N, Davis S J, (2016). Measuring Economic Policy Uncertainty[J]. Quarterly Journal

of Economics, 131: 1593–1636.

Barndorff-Nielsen O E, Shephard N, (2002). Estimating Quadratic Variation Using Realized Variance[J].

Journal of Applied Econometrics, 17: 457–477.

Barndorff-Nielsen O E, Shephard N, (2004). Econometric Analysis of Realized Covariation: High Fre-

quency Based Covariance, Regression, and Correlation in Financial Economics[J]. Econometrica, 72:

885–925.

Bierens H J, (1982). Consistent Model Specification Tests[J]. Journal of Econometrics, 20: 105–134.

Bierens H J, (1990). A Consistent Conditional Moment Test of Functional Form[J]. Econometrica, 58:

1443–1458.

Blundell R W, Powell J L, (2004). Endogeneity in Semiparametric Binary Response Models[J]. Review

of Economic Studies, 71: 655–679.

Bollerslev T, (1986). Generalized Autoregressive Conditional Heteroskedasticity[J]. Journal of Econo-

metrics, 31: 307–327.

Bollerslev T, Wooldridge J M, (1992). Quasi-maximum Likelihood Estimation and Inference in Dynamic

Models with Time-varying Covariances[J]. Econometric Reviews, 11: 143–172.

Breiman L, (2001). Statistical Modeling: The Two Cultures (with Comments and a Rejoinder by the

Author)[J]. Statistical Science, 16: 199–231.

Chen B, Hong Y, (2012). Testing for Smooth Structural Changes in Time Series Models via Nonpara-

metric Regression[J]. Econometrica, 80: 1157–1183.

Chen B, Hong Y, (2016). Detecting for Smooth Structural Changes in GARCH Models[J]. Econometric

Theory, 32: 740–791.

Chow G C, (1960). Tests of Equality between Sets of Coefficients in Two Linear Regressions[J]. Econo-

metrica, 28: 591–605.

Dahlhaus R, (1996). Maximum Likelihood Estimation and Model Selection for Locally Stationary Pro-

cesses[J]. Journal of Nonparametric Statistics, 6: 171–191.

Engle R F, (1982). Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of

United Kingdom Inflation[J]. Econometrica, 50: 987–1007.

Engle R F, Granger C W J, (1987). Co-integration and Error Correction: Representation, Estimation,

and Testing[J]. Econometrica, 55: 251–276.

Engle R F, Russell J R, (1998). Autoregressive Conditional Duration: A New Model for Irregularly

Spaced Transaction Data[J]. Econometrica, 66: 1127–1162.

Fan Y, Li Q, (1996). Consistent Model Specification Tests: Omitted Variables and Semiparametric

Functional Forms[J]. Econometrica, 64: 865–890.

Giannone D, Reichlin L, Small D, (2008). Nowcasting: The Real-time Informational Content of Macroe-

conomic Data[J]. Journal of Monetary Economics, 55: 665–676.

282 � �3 � 1�

Glosten L R, Jagannathan R, Runkle D E, (1993). On the Relation between the Expected Value and

the Volatility of the Nominal Excess Return on Stocks[J]. Journal of Finance, 48: 1779–1801.

Granger C W J, (1990). Modelling Economic Series: Readings in Econometric Methodology[M]. Oxford:

Oxford University Press, 369–383.

Granger C W, Newbold P, (1974). Spurious Regressions in Econometrics[J]. Journal of Econometrics,

2: 111–120.

Hall P, (1992). The Bootstrap and Edgeworth Expansion[M]. New York: Springer Science & Business

Media.

Hamilton J D, (1994). Time Series Analysis[M]. Princeton: Princeton University Press.

Han A, Hong Y, Wang S, (2017). Autoregressive Conditional Models for Interval-Valued Time Series

Data[R].Working Paper, Department of Economics, Cornell University.

Han A, Hong Y, Wang S, Yun X, (2016). A Vector Autoregressive Moving Average Model for Interval-

valued Time Series Data[C]// Essays in Hor of Aman Ullah Advances in Econometrics, 36: 417–460.

Hansen L P, (1982). Large Sample Properties of Generalized Method of Moments Estimators[J]. Econo-

metrica, 50: 1029–1054.

Hastie T, Tibshirani R, Wainwright M, (2015). Statistical Learning with Sparsity: The Lasso and

Generalizations[M]. Boca Raton: Taylor & Francis.

Hausman J A, (1978). Specification Tests in Econometrics[J]. Econometrica, 46: 1251–1271.

Heckman J J, Leamer E, (2001). Handbook of Econometrics[M]. Holland: Elsevier, 5: 3159–3228.

Hong Y, Lee Y J, (2013). A Loss Function Approach to Model Specification Testing and Its Relative

Efficiency[J]. Annals of Statistics, 41: 1166–1203.

Hong Y, Wang X, Wang S, (2017). Testing Strict Stationarity with Applications to Macroeconomic

Time Series[J]. International Economic Review, 58: 1227–1277.

Hong Y, White H, (1995). Constistent Specification Testing via Nonparametric Series Regression[J].

Econometrica, 63: 1133–1159.

Hsiao C, Ching S H, Wan S K, (2011). A Panel Data Approach for Program Evaluation: Measuring

the Benefits of Political and Economic Integration of Hong Kong with Mainland China[J]. Journal

of Applied Econometrics, 27: 705–740.

Imbens G W, Wooldridge J M, (2009). Recent Developments in the Econometrics of Program Evalua-

tion[J]. Journal of Economic Literature, 47: 5–86.

Jarque C M, Bera A K, (1980). Efficient Tests for Normality, Homoscedasticity and Serial Independence

of Regression Residuals[J]. Economics Letters, 6: 255–259.

Klein R W, Spady R H, (1993). An Efficient Semiparametric Estimator for Binary Response Models[J].

Econometrica, 61: 387–421.

Koenig E F, (2004). The Taylor Rule and the Transformation of Monetary Policy[M]. Stanford: Hoover

Press.

Lee S W, Hansen B E, (1994). Asymptotic Theory for the GARCH(1,1) Quasi-maximum Likelihood

Estimator[J]. Econometric Theory, 10: 29–52.

Lucas R E, (1976). Econometric Policy Evaluation: A Critique[C]// Carnegie-Rochester Conference

Series on Public Policy, 19–46.

Lumsdaine R L, (1996). Consistency and Asymptotic Normality of the Quasi-maximum Likelihood

Estimator in IGARCH(1,1) and Covariance Stationary GARCH(1,1) Models[J]. Econometrica, 64:

575–596.

� 2� ���: ����� � 283

Muller H G, (2005). Functional Modelling and Classification of Longitudinal Data[J]. Scandinavian

Journal of Statistics, 32: 223–240.

Muller H G, Stadtmuller U, (2005). Generalized Functional Linear Models[J]. Annals of Statistics, 33:

774–805.

Malkiel B G, Fama E F, (1970). Efficient Capital Markets: A Review of Theory and Empirical Work[J].

Journal of Finance, 25: 383–417.

Nelson C R, Plosser C R, (1982). Trends and Random Walks in Macroeconmic Time Series: Some

Evidence and Implications[J]. Journal of Monetary Economics, 10: 139–162.

Nelson D B, (1991). Conditional Heteroskedasticity in Asset Returns: A New Approach[J]. Economet-

rica, 59: 347–370.

Newey W K, (1985). Generalized Method of Moments Specification Testing[J]. Journal of Econometrics,

29: 229–256.

Newey W K, West K D, (1987). A Simple, Positive Semi-definite, Heteroskedasticity and Autocorrela-

tion Consistent Covariance Matrix[J]. Econometrica, 55: 703–708.

Newey W K, West K D, (1994). Automatic Lag Selection in Covariance Matrix Estimation[J]. Review

of Economic Studies, 61: 631–653.

Noureldin D, Shephard N, Sheppard K, (2011). Multivariate High-frequency-based Volatility (HEAVY)

Models[J]. Journal of Applied Econometrics, 27: 907–933.

Phillips P C B, (1977a). An Approximation to the Finite Sample Distribution of Zellner’s Seemingly

Unrelated Regression Estimator[J]. Journal of Econometrics, 6: 147–164.

Phillips P C B, (1977b). Approximations to Some Finite Sample Distributions Associated with a First-

order Stochastic Difference Equation[J]. Econometrica, 45: 463–485.

Phillips P C B, (1977c). A General Theorem in the Theory of Asymptotic Expansions as Approximations

to the Finite Sample Distributions of Econometric Estimators[J]. Econometrica, 45: 1517–1534.

Phillips P C B, (1986). Understanding Spurious Regressions in Econometrics[J]. Journal of Economet-

rics, 33: 311–340.

Phillips P C B, (1987a). Time Series Regression with a Unit Root[J]. Econometrica, 55: 277–301.

Phillips P C B, (1987b). Towards a Unified Asymptotic Theory for Autoregression[J]. Biometrika, 74:

535–547.

Ramsay J, Silverman B W, (2002). Applied Functional Data Analysis: Methods and Case Studies[M].

New York: Springer.

Ramsay J, Silverman B W, (2005). Functional Data Analysis[M]. New York: Springer.

Robinson P M, (1988). Root-N-Consistent Semiparametric Regression[J]. Econometrica, 56: 931–954.

Shephard N, Sheppard K, (2010). Realising the Future: Forecasting with High-frequency-based Volatil-

ity (HEAVY) Models[J]. Journal of Applied Econometrics, 25: 197–231.

Staiger D, Stock J H, (1997). Instrumental Variables Regression with Weak Instruments[J]. Economet-

rica, 65: 557–586.

Stock J H, Trebbi F, (2003). Retrospectives: Who Invented Instrumental Variable Regression?[J].

Journal of Economic Perspectives, 17: 177–194.

Sun Y, Han A, Hong Y, Wang S, (2018). Threshold Autoregressive Models for Interval-valued Time

Series Data[J]. Journal of Econometrics, 206: 414–446.

Sun Y, Hong Y, Wang S, (2019). Out-of-Sample Forecasts for China’s Economic Growth and Inflation

Using Rolling Weighted Least Squares[J]. Journal of Management and Engineering, 4(1): 1–11.

284 � �3 � 1�

Tauchen G, (1985). Diagnostic Testing and Evaluation of Maximum Likelihood Models[J]. Journal of

Econometrics, 30: 415–443.

Ullah A, (1990). Finite Sample Econometrics: A Unified Approach[M]. New York: Springer.

Varian H R, (2014). Big Data: New Tricks for Econometrics[J]. Journal of Economic Perspectives, 28:

3–28.

Wang S, Yu L, Lai K K, (2005). Crude Oil Price Forecasting with TEI@I Methodology[J]. Journal of

Systems Science and Complexity, 18: 145–166.

White H, (1980). A Heteroskedasticity-consistent Covariance Matrix Estimator and a Direct Test for

Heteroskedasticity[J]. Econometrica, 48: 817–838.

White H, (1981). Consequences and Detection of Misspecified Nonlinear Regression Models[J]. Journal

of the American Statistical Association, 76: 419–433.

White H, (1984). Asymptotic Theory for Econometricians[M]. San Diego: Academic Press.

White H, (1992). Artificial Neural Networks: Approximation and Learning Theory[M]. Oxford: Black-

well Publishers, Inc.

White H, Stinchcombe M, (1991). Adaptive Efficient Weighted Least Squares with Dependent Obser-

vations[M]. New York: Springer.