기업이질성에 근거한 수출 참여 예측: 딥러닝을 이용한 시계열 분류 = Prediction of Export Participation Based on Firm Heterogeneity: Time Series Classification with Deep Learning

다국어 초록 (Multilingual Abstract)

Purpose: This study is to predict firms’ export participation based on firm heterogeneity, considering the situation where many countries around the world try to promote firms’ entry to export markets from the perspective of heterogeneous firm trade framework these days.
Research design, data, and methodology: We used the 13-year time series data from the business activity survey produced by the Statistics Korea. Total factor productivity, financial leverage, and R&D expenditure were used as input variables and export participation was used as an output variable for time series classification with deep learning. We have trained and compared the three deep learning models for time series classification: multi layer perceptron, fully convolutional network, and residual network. We implemented the models using the open source deep learning library Keras with the Tensorflow back-end. The models’ performance was evaluated using the mean of accuracy, precision, recall, and F1-score over the 10 runs on the testing data set.
Results: The results showed that the fully convolutional network (FCN) architecture performs best for the time series classification task and the recall is higher than the precision. The accuracy of the best model is 0.86, the precision is 0.64, the recall is 0.80, and the F1-score is 0.71.
Conclusions: This study contributes to promoting the understanding of deep learning approach to prediction of export participation in the context of heterogenous firm trade theory. The prediction focuses on the selection of non-exporting firms, from the perspective of policy orientation for excavating and making firms without exporting start exporting. We propose to be able to utilize the FCN for enhancing the effectiveness and efficiency of export promotion policies, in particular focused on increase in firms’ export participation, by interpreting three of the indicators being used for model evaluation, precision, recall, and F1-score, in the context of such policy.

참고문헌 (Reference)

1 METI(Ministry of Economy, Trade and Industry), "White Paper on International Economy and Trade 2017" Trade Policy Bureau, METI 2017

2 Zheng, Y., "Web-Age Information Management" Springer 298-310, 2014

3 Haidar, J. I., "Trade and productivity : Self-selection or Learning-by-exporting in India" 29 (29): 1766-1773, 2012

4 Kiriyama, N., "Trade and Innovation: Synthesis Report" OECD Publishing 2012

5 Cernat, L., "Towards"Trade Policy Analysis 2. 0"From National Comparative Advantage to Firm-level Trade Data" 4 : 1-12, 2014

6 Serrà, J., "Towards a Universal Neural Network Encoder for Time Series" 120-129, 2018

7 Wang, Z., "Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline" 1578-1585, 2017

8 Tanisaro, P., "Time Series Classification Using Time Warping Invariant Echo State Networks" IEEE 831-836, 2016

9 Costantini, J., "The Organization of Firms in a Global Economy" Harvard University Press 15-36, 2007

10 Melitz, M. J., "The Impact of Trade on Intra-industry Reallocations and Aggregate Industry Productivity" 71 (71): 1695-1725, 2003

11 Broocks, A., "The Impact of Export Promotion on Export Market En" 107 : 19-33, 2017

12 Ge, J., "The Effects of GVC Embeddendness on Productivity Improvement: From the Perspective of R&D and Government Subsidy" 135 : 22-31, 2018

13 Roberts, M. J., "The Decision to Export in Colombia : An Empirical Mode of Entry with Sunk Costs" 87 (87): 545-564, 1997

14 Serti, F., "Self Selection among Different Export Markets" 117 (117): 102-105, 2012

15 Bhagwati, J., "Reforms and Economic Transformation in india" Oxford University Press 2012

16 Lapham, B., "Redesigning Canadian Trade Policies for New Global Realities. The Art of the State Series, Volume VI" Institute for Research on Public Policy 39-73, 2017

17 Yang, C. H., "R&D, Productivity, and Exports : Plant-level Evidence from Indonesia" 29 (29): 208-216, 2012

18 Kasahara, H., "Productivity and the Decision to Import and Export : Theory and Evidence" 89 (89): 297-316, 2013

19 Ciuriak, D., "Policy Implications of Heterogeneous Firms Trade Theory" 2013

20 Bernard, A. B., "Plants and Productivity in International Trade" 93 (93): 1268-1290, 2003

21 LeCun, Y. A., "Neural Networks: Tricks of the Trade" Springer 9-48, 2012

22 Cui, Z., "Multi-scale Convolutional Neural Networks for Time Series Classification"

23 Costinot, A., "Micro to Macro: Optimal Trade Policy with Firm Heterogeneity" NBER 2016

24 Foster-McGregor, N., "Learning-by-exporting versus Self-selection : New Evidence for 19 Sub-Saharan African Countries" 125 (125): 212-214, 2014

25 Sharma, C., "International Trade and Performance of Firms : Unraveling Export, Import and Productivity Puzzle" 57 : 61-74, 2015

26 Deng, Z., "Innovation and Survival of Exporters : A Contingency Perspective" 23 (23): 396-406, 2014

27 Geldres-Weis, V. V., "Innovation and Experiential Knowledge in Firm Exports : Aplying the Initial U-model" 69 (69): 5076-5081, 2016

28 Manova, K., "How Firms Export: Processing vs. Ordinary Trade with Financial Frictions" 100 : 120-137, 2016

29 Plouffe, M., "Heterogeneous Firms and Trade-Policy Stances : Evidence from a Survey of Japanese Producers" 19 (19): 1-40, 2017

30 Kim, H. S., "Firms’ Leverage and Export Market Participation: Evidence from South Korea" 148 : 41-58, 2016

31 Kim, I. S., "Firms in Trade and Trade Politics" 22 : 399-417, 2019

32 Ciuriak, D., "Firms in International Trade : Trade Policy Implications of the New New Trade Theory" 6 (6): 130-140, 2015

33 Caggese, A., "Financing Constraints, Firm Dynamics, Export Decisions, and Aggregate Productivity" 16 (16): 177-193, 2013

34 Chan, J. M. L., "Financial Frictions and Trade Intermediation: Theory and Evidence" 119 : 567-593, 2019

35 Berman, N., "Financial Factors and the Margins of Trade : Evidence from Cross-country Firm-level Data" 93 (93): 206-217, 2010

36 Greenway, D., "Financial Factors and Export Decisions" 73 (73): 377-395, 2007

37 Bellone, F., "Financial Constraints and Firm Export Behaviour" 33 (33): 347-373, 2010

38 Nagaraj, P., "Financial Constraints and Export Participation in India" 140 : 19-35, 2014

39 Dzhumashev, R., "Exporting, R&D Investment and Firm Surival in the Indian IT Sector" 42 : 1-19, 2016

40 Bernard, A. B., "Export Entry and Exit by German Firms" 137 : 105-123, 2001

41 Bernard, A. B., "Exceptional Exporter Performance: Cause, Effect, or Both?" 47 (47): 1-25, 1999

42 Levinsohn, J., "Estimating Production Functions using Inputs to Control for Unobservables" 70 (70): 317-342, 2003

43 Montalbano, P., "Energy Efficiency, Productivity and Exporting: Firm-level Evidence in Latin America" 79 : 97-110, 2017

44 Liu, B. J., "Do Bigger and Older Firms Learn and More from Exporting? Evidence from China" 45 : 89-102, 2017

45 Myers, S., "Determinants of Corporate Borrowing" 5 (5): 147-175, 1977

46 Papernot, N., "Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning"

47 Fawaz, H. I., "Deep Learning for Time Series Classification : A Review" 33 (33): 917-963, 2019

48 Le Guennec, A, "Data Augmentation for Time Series Classification Using Convolutional Neural Networks"

49 Manova, K., "Credit Constraints, Heterogeneous Firms, and International Trade"

50 Manova, K., "Credit Constraints, Heterogeneous Firms and International Trade" 80 (80): 711-744, 2013

51 Fauceglia, D., "Credit Constraints, Firm Exports and Financial Development: Evidence from Developing Countries" 55 : 53-66, 2015

52 DiPietro, W. R., "Creativity, Innovation, and Export Performance" 26 (26): 133-139, 2006

53 Geng, Y., "Cost-Sensitive Convolution Based Neural Networks for Imbalanced Time-Series Classification"

54 Myers, S., "Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have" 13 (13): 187-221, 1984

55 Zhao, B., "Convolutional Neural Networks for Time Series Classification" 28 (28): 162-169, 2017

56 Jensen, M. C., "Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers" 76 (76): 323-329, 1986

57 Kingma, D. P., "Adam: A method for stochastic optimization"

58 Zeiler, M. D., "Adadelta: an adaptive learning rate method"