Jordan neural network for inflation forecasting

Abstract

In times of pronounced nonlinearity of macroeconomic variables and in situations when variables are not normally distributed, i.e. when the assumption of i.i.d. is not fulfilled, neural networks (NNs) should be used for forecasting. In this paper, Jordan neural network (JNN), a special type of NNs is examined, because of its advantages in time series forecasting suitable for inflation forecasting. The variables used as inputs include labour market variable, financial variable, external factor and lagged inflation, i.e. the most commonly used variables in previous researches. The research is conducted at the aggregate level of euro area countries in the period from January 1999 to January 2017. Based on 250 estimated JNNs, which differ in selected variables, sample breaking point and varying parameters (number of hidden neurons, weight value of the context unit), the model adequacy indicators for each JNN are calculated for two periods: in-the-sample and out-of-sample. Finally, the optimal JNN for inflation forecasting is obtained as the best compromise solution between low mean squared error inthe-sample and out-of-sample and low number of parameters to estimate. This paper contributes to existing literature in using JNN for inflation forecasting since it is rarely used for macroeconomic time series prediction in general. Moreover, this paper defines which set of variables contributes to the best inflation forecast. Additionally, JNN is examined thoroughly by fixing certain parameters of the model and alternating other parameters to contribute to the JNN literature, i.e. finding the optimal JNN.

Can Recurrent Neural Networks Predict Inflation in Euro Zone as Good as Professional Forecasters?

“…These findings support our modeling strategy when neural networks are considered in practical applications, i.e., for forecasting purposes. This is in line with [18,38] who found that multivariate models and more variables can improve inflation forecasts, while it is opposite to findings of [24,28] favoring simpler models. The same conclusion emerges when JNNs are compared with appropriate ARIMA models.…”

“…Ref. [24] compared JNN only with different exogenous inputs to conclude that in most cases the simplest JNNs are ranked the highest by their performance, i.e., JNNs with lagged dependent variable and one exogenous regressor. Ref.…”

Machine learning prediction of agricultural produces for Indian Farmers using LSTM

“…The classical Box-Jenkins Models only work with linear data, whereas the machine learning models (RNN) work well on a wide range of data including nonlinear data. Furthermore, as per the research regarding the prediction of agriculture commodities prices in developing countries like India [13][14][15][16][17][18][19] , Recurrent Neural Network is specified best for nonlinear, adaptable, and strongly mapped data. Therefore, we will be using the LSTM (special recurrent neural network) model, over other models like the ARIMA model, for predicting the agriculture commodity price.…”

What factors influence Bitcoin’s daily price direction from the perspective of machine learning classifiers?

“…Similarly, Poyser [22] has not found any relevant effect of confirmation time, hash rate, and the number of transactions per day on Bitcoin's price, which is contrary to results from [16] and [9] that show a positive, albeit small, effect of blockchain variables on Bitcoin's price. When it comes to sentiment analysis measures, most researchers agree that attractiveness is the main driver of Bitcoin price ( [22], [10], [24], [23]). Even though there is no consensus regarding the impact of macro-finance factors on Bitcoin prices, the macro-finance variables most usually associated with Bitcoin prices are the S&P500, gold, oil, real estate, VIX, and exchange rates.…”