At the recent R/Finance 2014 conference in Chicago I gave a talk on Smooth Transition AR models and a new package for estimating them called twinkle. In this blog post I will provide a short outline of the models and an introduction to the package and its features. Financial markets ... [Read more...]

The last model added to the rugarch package dealt with the modelling of intraday volatility using a multiplicative component GARCH model. The newest addition is the realized GARCH model of Hansen, Huang and Shek (2012) (henceforth HHS2012) which relates the realized volatility measure to the latent volatility using a flexible representation ... [Read more...]

What is risk parity (RP)? Simply put, it is a method of allocating equal risk shares to each asset in the portfolio. In more traditional allocation schemes, equity, being the riskiest asset (and hence providing the highest reward), has typically received the lion’s share. With RP, equalization of risk ... [Read more...]

The Independent Factor Autoregressive Conditional Density (IFACD) model of Ghalanos, Rossi and Urga (2014) uniquely, in its class of parametric models, generates time varying higher co-moment forecasts, as a consequence of the ACD specification of the conditional density of the standardized innovations. In this short note I discuss in more detail ... [Read more...]

In the global economic landscape, national borders have become increasingly blurred. Local economies depend on competitively priced global inputs and on well functioning and prosperous global markets for their exports. This interdependency also means that as a system, a crisis in one area quickly spreads to others areas like a ... [Read more...]

In the previous blog article I discussed a dynamic binary model for the directional forecast of the US equity market using a select number of economic, fundamental and technical variables as predictors. A natural direction for extending that research would be to look at similar models in different countries or ... [Read more...]

While it is generally accepted that the returns of financial assets are almost impossible to forecast with any degree of accuracy which would provide meaningful profit1 , there is evidence that the sign of the returns is much more forecastable. Theoretically, Christoffersen and Diebold (2006) have shown how the forecastability of the ... [Read more...]

This is a quick update regarding the status of my R packages on google code. Since google decided to disallow uploads from Jan-2014 for existing projects, and immediately for new ones (meaning that the tarballs and zips could not be hosted on their servers anymore), I have had no choice ... [Read more...]

In the last few years there has been a increasing tendency to ignore the value of a disciplined quantitative approach to the portfolio allocation process in favor of simple and static weighting schemes such as equal weighting or some type of adjusted volatility weighting. The former simply ignores the underlying ... [Read more...]

The Autoregressive Conditional Density (ACD) model of Hansen (1994) extended GARCH models to include time variation in the higher moment parameters. It was a somewhat natural extension to the premise of time variation in the conditional mean and variance, though it probably raised more questions than it, or subsequent research have ... [Read more...]

The interest in high frequency trading and models has grown exponentially in the last decade. While I have some doubts about the validity of any signals emerging from all the noise at higher and higher frequencies, I have nevertheless decided to look at the statistical modelling of intraday returns using ... [Read more...]

Since the last release of rugarch on CRAN (ver 1.0-16), there have been many changes and new features in the development version of the package (ver 1.01-5). First, development of the package (and svn) has been moved to google code from r-forge. Second, the package now features exclusive use of ... [Read more...]

In their paper on GARCH model comparison, Hansen and Lunde (2005) present evidence that among 330 different models, and using daily data on the DM/$ rate and IBM stock returns, no model does significantly better at predicting volatility (based on a realized measure) than the GARCH(1,1) model, for an out of sample ... [Read more...]

This short demonstration illustrates the use of the DCC model and its methods using the rmgarch package, and in particular an alternative method for 2-stage DCC estimation in the presence of the MVT distribution shape (nuisance) parameter. The theoretical background and representation of the model is detailed in the package’... [Read more...]