id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_ic4cnnv7svfkboukympxr6ybyu	Yuan Yao	Data analytics enhanced component volatility model	2017	28	.pdf	application/pdf	9082	1051	73	The volatility data is decomposed by the lowpass filter into long and short term components, which are then modelled by the autoregressive neural network and then model the long-term component using an autoregressive neural network and the short-term (ARNN), a special format of artificial neural network, to model and forecasting the long-term component 𝐿𝑡 forecasting the long-term component of volatility, ARNN operates as a non-linear regression function at the long-term component 𝐿𝑡 of the realized volatility, an autoregressive neural network (ARNN) with three We can obtain the forecasted the long-term component �̂�𝑡+𝑛 of the realized volatility through ARNN model. We use the neural network enhanced two-component model defined in Section 2 to forecasting the volatility autoregressive neural network enhanced two-component model constructed by one hidden layer with 10 neurons (NNE2C 10 autoregressive neural network enhanced two-component model constructed by one hidden layer with 10 neurons (NNE2C 10 	./cache/work_ic4cnnv7svfkboukympxr6ybyu.pdf	./txt/work_ic4cnnv7svfkboukympxr6ybyu.txt