Combining robust dynamic neural networks with traditional technical indicators for generating mechanic trading signals

Abstract

Forecasting assets’ prices is the aim of each trader, although the trading approaches employed may vary a lot. The development of machine learning techniques has brought the opportunity to design mechanic trading systems based on dynamic artificial neural networks. The aim of this paper is to combine traditional technical indicators [such as exponential weighted moving average (EWMA), percentage volume oscillator (PVO) and stochastic indicator — %K and %D] with the nonlinear autoregressive networks (NAR and NARX). The first part of the paper describes how neural networks designed for forecasting time series work, the second one performs a deeper validation of the code and the third one combines the dynamic networks with traditional technical indicators in order to generate reliable mechanic signals. The article ends with a back testing of the trading system performed on Dow Jones Industrial Average and on Nasdaq Composite Indexes.

Classification of NPL with a Random Forest approach

“…where the exposure of the given dossier selected by the recovery event (i) at granular level -a non-linear function of the estimated probability of recovery -is multiplied by the estimated recovery rate for the given dossier ( ) . This is different from the average recovered amount calculated through the average recovery rate over the given perimeter (10) ER ̅̅̅̅ = ∑ * ∈ real rec…”

“…Uncertainty is represented using scenario trees, building on a long tradition of multi-period stochastic models that find numerous applications in the risk management of financial institutions. 10…”

NPL disposal treatment in the LGD estimates

“…In particolare, visti i risultati, si decide di escludere dalle successive analisi i settori automotive, dell'industria chimica, dell'intrattenimento, della produzione industriale e finanziario. Selezionati a partire dal campione complessivo gli indici risultati significativi in termini di asset allocation, si procede, per tale sottoinsieme, a calcolare le proiezioni impiegando come strumento di forecasting le reti neurali dinamiche di tipo NAR (Non-Linear Auto Regressive) e NARX (Non-Linear Auto Regressive with Exogenous Variables) [10].…”

“…dove viene introdotto il numero di ritardi da applicare sia alla variabile endogena (il livello dei prezzi), , sia alla variabile esogena (ad esempio i volumi), [10].…”

Non-Performing Exposures of banks: impatient diktats and national solutions vs patient management and Asset Management Companies at European level

“…11 9 These estimates were obtained running Fama-MacBeth cross-section regressions on the panel data in our sample of the yields of 10-year government bonds on the ratings of political stability and economic policy confidence. 10 See, for instance, Mulvey, J. and W.T. Ziemba A significant short term political risk for DSA is the fiscal stance of the new government following the 2018 elections.…”