Code Sys.setenv ( TZ = "UTC" ) # load required packages library (iTensor) library (tsmarch) library (flextable) library (data.table) library (xts) Motivation The Generalized Orthogonal GARCH (GOGARCH) model uses Independent Component Analysis (ICA) to extract independent sources of volatility from multivariate time series, enabling more precise and interpretable modeling of their joint dynamic…
Code # load required packages library (tsmethods) library (tsaux) library (tsets) library (xts) library (tsdistributions) library (data.table) library (future.apply) library (magrittr) library (ggplot2) Motivation Finding stable patterns in data is a key requirement if we are to have any success in forecasting the future. These patterns are identified by analyzing the historical behavior of data,…
Motivation Real world data rarely has clean patterns from which we can quickly identify the data generating process (DGP). It is messy 1 and contains anomalies which often times cloud the true signal. Even worse, even if the data appears regular, there are never any guarantees that the future will follow the past. Choosing a best model is an almost impossible task. At best, a model we choose…
Code library (tsdistributions) library (data.table) assignInNamespace ( "cedta.override" , c (data.table ::: cedta.override, "distill" ), "data.table" ) library (ggplot2) library (gridExtra) f1 <- function (x) ddist ( "ghst" , x, 2 , 1 , skew = - 20 , shape = 10 ) f2 <- function (x) ddist ( "ghst" , x, - 2 , 1 , skew = 20 , shape = 10 ) ggplot ( NULL , aes ( c ( - 6 , 6 ))) + geom_area ( stat =…
The tsdistributions package provides the function authorized_domain which calculates the region of Skewness-Kurtosis for which a density exists. This is related to the Hamburger moment problem, and the maximum attainable Skewness (S) given kurtosis (K) (see (Widder 2015) ). A key takeaway from Widder is that for as given level of kurtosis( ) only a finite range of skewness ( ) may be attained,…