|
on Forecasting |
| By: | Pesaran, M. H.; Pick, A.; Timmermann, A. |
| Abstract: | We develop novel forecasting methods for panel data with heterogeneous parameters and examine them together with existing approaches. We conduct a systematic comparison of their predictive accuracy in settings with different cross-sectional (N) and time (T) dimensions and varying degrees of parameter heterogeneity. We investigate conditions under which panel forecasting methods can perform better than forecasts based on individual estimates and demonstrate how gains in predictive accuracy depend on the degree of parameter heterogeneity, whether heterogeneity is correlated with the regressors, the goodness of fit of the model, and, particularly, the time dimension of the data set. We propose optimal combination weights for forecasts based on pooled and individual estimates and develop a novel forecast poolability test that can be used as a pretesting tool. Through a set of Monte Carlo simulations and three empirical applications to house prices, CPI inflation, and stock returns, we show that no single forecasting approach dominates uniformly. However, forecast combination and shrinkage methods provide better overall forecasting performance and offer more attractive risk profiles compared to individual, pooled, and random effects methods. |
| Keywords: | Forecasting, Panel data, Heterogeneity, Forecast evaluation, Forecast combination, Shrinkage, Pooling |
| JEL: | C33 C53 |
| Date: | 2022–03–21 |
| URL: | http://d.repec.org/n?u=RePEc:cam:camdae:2219&r= |
| By: | Daniel Hopp |
| Abstract: | The COVID-19 pandemic has demonstrated the increasing need of policymakers for timely estimates of macroeconomic variables. A prior UNCTAD research paper examined the suitability of long short-term memory artificial neural networks (LSTM) for performing economic nowcasting of this nature. Here, the LSTM's performance during the COVID-19 pandemic is compared and contrasted with that of the dynamic factor model (DFM), a commonly used methodology in the field. Three separate variables, global merchandise export values and volumes and global services exports, were nowcast with actual data vintages and performance evaluated for the second, third, and fourth quarters of 2020 and the first and second quarters of 2021. In terms of both mean absolute error and root mean square error, the LSTM obtained better performance in two-thirds of variable/quarter combinations, as well as displayed more gradual forecast evolutions with more consistent narratives and smaller revisions. Additionally, a methodology to introduce interpretability to LSTMs is introduced and made available in the accompanying nowcast_lstm Python library, which is now also available in R, MATLAB, and Julia. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.11872&r= |
| By: | Yuxuan Huang; Luiz Fernando Capretz; Danny Ho |
| Abstract: | Application of machine learning for stock prediction is attracting a lot of attention in recent years. A large amount of research has been conducted in this area and multiple existing results have shown that machine learning methods could be successfully used toward stock predicting using stocks historical data. Most of these existing approaches have focused on short term prediction using stocks historical price and technical indicators. In this paper, we prepared 22 years worth of stock quarterly financial data and investigated three machine learning algorithms: Feed-forward Neural Network (FNN), Random Forest (RF) and Adaptive Neural Fuzzy Inference System (ANFIS) for stock prediction based on fundamental analysis. In addition, we applied RF based feature selection and bootstrap aggregation in order to improve model performance and aggregate predictions from different models. Our results show that RF model achieves the best prediction results, and feature selection is able to improve test performance of FNN and ANFIS. Moreover, the aggregated model outperforms all baseline models as well as the benchmark DJIA index by an acceptable margin for the test period. Our findings demonstrate that machine learning models could be used to aid fundamental analysts with decision-making regarding stock investment. |
| Date: | 2022–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2202.05702&r= |