A neural network ensemble approach for GDP forecasting

Longo, Luigi; Riccaboni, Massimo; Rungi, Armando

doi:10.1016/j.jedc.2021.104278

Cited by 29 publications

(13 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning has an enormous potential to enhance the quality of the prediction of economic growth and competitiveness (Longo et al 15 ). In the present work, we applied Matrix Completion (MC) to investigate the economic complexity of countries in various ways .…”

Section: Discussionmentioning

confidence: 99%

A machine learning approach to economic complexity based on matrix completion

Gnecco

Nutarelli

Riccaboni

2022

Sci Rep

Self Cite

View full text Add to dashboard Cite

This work applies Matrix Completion (MC) – a class of machine-learning methods commonly used in recommendation systems – to analyze economic complexity. In this paper MC is applied to reconstruct the Revealed Comparative Advantage (RCA) matrix, whose elements express the relative advantage of countries in given classes of products, as evidenced by yearly trade flows. A high-accuracy binary classifier is derived from the MC application to discriminate between elements of the RCA matrix that are, respectively, higher/lower than one. We introduce a novel Matrix cOmpletion iNdex of Economic complexitY (MONEY) based on MC and related to the degree of predictability of the RCA entries of different countries (the lower the predictability, the higher the complexity). Differently from previously-developed economic complexity indices, MONEY takes into account several singular vectors of the matrix reconstructed by MC. In contrast, other indices are based only on one/two eigenvectors of a suitable symmetric matrix derived from the RCA matrix. Finally, MC is compared with state-of-the-art economic complexity indices, showing that the MC-based classifier achieves better performance than previous methods based on the application of machine learning to economic complexity.

show abstract

Section: Discussionmentioning

confidence: 99%

A machine learning approach to economic complexity based on matrix completion

Gnecco

Nutarelli

Riccaboni

2022

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…We chose a subset of the available variables to use in our models for predicting GDP growth rate. This included GDP, net population, CPI, employment rate, share of labour compensation in GDP at current national prices, exchange rate, gross domestic product per capital (CGDPo), and price-levels that refer to the perspectives of macroeconomics (production, expenditure, and trade), finance, and human resources, supplemented by a price index that can show whether inflation is present, to help us better predict economic development and economic crises [114,105,83,71]. We transform data into year-on-year ratio in order to predict GDP growth rate.…”

Section: Datamentioning

confidence: 99%

“…and soft indicators which is a less tangible community characteristics and values [94] (confidence index, exchange rate, etc.). Longo et al [71] used the ensemble method based on the LSTM to forecast the GDP of the United States and provided insights about the contribution of different features (economic indicators) during COVID-19. The economic indicators typically used for GDP forecasting include 141 variables taken form the FRED database (US Federal reserve Database), see also [43].…”

Section: Introductionmentioning

confidence: 99%

Recursive deep learning framework for forecasting the decadal world economic outlook

Wang¹,

Beard²,

Hawkins³

et al. 2023

Preprint

View full text Add to dashboard Cite

Gross domestic product (GDP) is the most widely used indicator in macroeconomics and the main tool for measuring a country's economic ouput. Due to the diversity and complexity of the world economy, a wide range of models have been used, but there are challenges in making decadal GDP forecasts given unexpected changes such as pandemics and wars. Deep learning models are well suited for modeling temporal sequences have been applied for time series forecasting. In this paper, we develop a deep learning framework to forecast the GDP growth rate of the world economy over a decade. We use Penn World Table as the source of our data, taking data from 1980 to 2019, across 13 countries, such as Australia, China, India, the United States and so on. We test multiple deep learning models, LSTM, BD-LSTM, ED-LSTM and CNN, and compared their results with the traditional time series model (ARIMA,VAR). Our results indicate that ED-LSTM is the best performing model. We present a recursive deep learning framework to predict the GDP growth rate in the next ten years. We predict that most countries will experience economic growth slowdown, stagnation or even recession within five years; only China, France and India are predicted to experience stable, or increasing, GDP growth.

show abstract

“…With the continuous development and optimization of machine learning, neural networks have been widely used in prediction, mainly including convolutional neural networks (CNN) [7] and recursive neural networks (RNN) [8] [9]. In addition, there are also some other methods applied in the financial field, such as XGBoost [10] and AdaBoost algorithms [11], both achieved high prediction accuracy. In a word, machine learning is applied in economic analysis field abundantly.…”

Section: Related Workmentioning

confidence: 99%

Predicting Impact of COVID-19 on the Global Economy Based on Hybrid Model

Zuo¹,

Ma²,

Chen³

2022

Proceedings of the 2022 7th International Conference on Financial Innovation and Economic Development (ICFIED 2022)

View full text Add to dashboard Cite

The outbreak of COVID-19 has caused hitherto unknown damage to the global economy, resulting in a large number of company failures and staff unemployment. In order to suppress the further spread of the virus, many countries and regions have adopted quarantine policies to varying degrees. This has led to the failure of normal business activities and financial market behaviors. The health crisis caused by the new crown pneumonia has gradually turned into an economic crisis. In this paper, the data provided by the KAGGLE competition platform is used to analyze and predict the impact of COVID-19 on the economy. A mixed prediction model (AdaBoost, Linear Regression, Decision Tree) is constructed to predict the GDP per capita of different countries. The results show that the hybrid model has the highest accuracy. Specifically, the lowest MSE index score of our model is 7.57, and KNN and decision tree algorithm are 2.23 and 2.19 higher than our hybrid model respectively.

show abstract

A neural network ensemble approach for GDP forecasting

Cited by 29 publications

References 35 publications

A machine learning approach to economic complexity based on matrix completion

A machine learning approach to economic complexity based on matrix completion

Recursive deep learning framework for forecasting the decadal world economic outlook

Predicting Impact of COVID-19 on the Global Economy Based on Hybrid Model

Contact Info

Product

Resources

About