... the Highly Acclaimed Graphics Tool (Free on-line Trial...)

Horizon II is used by traders, marketers, financial engineers and risk managers in banks, brokers and financial institutions - over 650 in 14 countries around the world. It provides the user with a rapid facility to carry out extensive study of the inter-bank yield curve, its components and derivatives.

Supported by a database of over 60 currencies, Horizon provides easy-to-use facilities to analyse the market so that more informed trading decisions can be made. It has the unique ability to present, on one display, forward rates the market was forecasting on any particular day and the actual rates that pertained in the market. Some examples of how Horizon can be used are given below.

One of the main attributes of Horizon is the ease and speed with which it can be used. A user can test out theories in seconds and thus take advantage of market opportunities. He can analyse not only basic data, but also derived information such as zero coupon and forward curves. Rate spread, correlation, volatility, moving averages and regression analyses can be accessed both within and across markets. If required, interest-rate Implied Volatility Data (caps, floors and swaptions) can also be accessed. A powerful graph definition dialogue enables users to write their own expressions, using rate series as variables, to test alternative strategies and to save expressions for future use.

The quality of data in the underlying database is regarded by our customers as the best available in the marketplace. Collected from multiple sources, at the close of the futures market each day, all data is statistically and manually checked before the database is updated. Term structures are constructed daily from cash and swaps data, going back to 1986 for the major currencies. Market indices, F/X spot, and Bond data is also collected. 

PRODUCT OVERVIEW
Creating a Graph in Horizon
Spread of Forward Rates
Implied Volatility
Analysing Market Predictions of Interest rates
Volatility Analysis
Adjustment of Swap Rate Compounding
Term Structure Spreads
Combining Term Structures with Time Series

Click for Rolling Demo

Creating a Graph in Horizon

All graphs are created using the ‘Graph Definition Dialogue’ (GDD) box, which is permanently attached to a graph for future reference and amendment. The GDD is divided into three sections. The upper section is used to select a curve, the middle section is where a curve is edited, and the lower section holds the curves to be displayed and the time period covered by the graph.

The upper section assists the user by dynamically displaying options, in the form of drop-down menus, depending on selections made from each previous menu.

The middle section is the Edit Box. Curves are built here when using the special functions of Horizon.

The lower section holds the finalised curves and information on the axis to be used together with the date range of the display.

Back to top of page

Spread of Forward Rates

This graph represents the spread between the USD and the GBP 15 year swap in 15 years time. The rate is displayed on the left hand axis and the spread on the right hand axis.

Back to top of page

Implied Volatility

Caps and Floors at the money

Swaptions

This graph represents the volatility at the money of a 5 year option on a 5 year swap.

Back to top of page

Analysing Market Predictions of Interest Rates

The Shift function displaces results earlier or later along the time axis, enabling comparison of events where some form of time-lag effect is postulated. Correlation is a measure of how well the best relationship of a specified kind fits a set of data.

In this example we show an example of the inability of the market to predict interest rates in the future. We have plotted the correlation of the DEM 3m/6m forward series with the three months cash series shifted back three months. We have used the correlation as a measure of the success (or lack of success) of the market to predict cash rates from the forward curve.

Back to top of page

Volatility Analysis

The graph below represents the volatility of a Sterling two year swap during the exit of Sterling from the ERM system (1992), using three different methods available within HorizonII. While the annualised standard deviation of returns is the method most commonly used by many practitioners, it has the problem that it generates a sustained high volatility figure for the entire sampling period being used if there is a large movement in the underlying rates, as can be clearly seen in the graph.

In practice, implied volatility will tend to revert to a previously established mean or to a slightly different level if fundamentals of the product have changed significantly.

For this reason, we have implemented the GARCH and Moving Average methods. The GARCH model has been chosen as it is one of the models becoming increasingly used by practitioners and one that benefits from the concept of mean reversion. One of the pitfalls with this model is evident when events like the Sterling ERM exit occurs. The volatility calculated by the GARCH model on this date is extremely high and from comparisons done with implied volatilities is disproportionate to market implied volatilities.

In order to compensate for this weakness in the GARCH model we have implemented a redesign of GARCH which we have called the Moving Average Model. This behaves in very much the same fashion as the GARCH model except in times of extreme volatility when it tends to not overstate the numbers in the same way as the GARCH.

The parameters chosen for each of these models were designed to give a high degree of correlation with futures options implied volatilities.

Back to top of page

Adjustment of Swap Rate Compounding

In this example, we are looking at the spread between a sterling 2 years swap (Act/365, semi annualised coupon) and a USD 2 year swap (Act/360, annualised coupon). Using the compounding function and the edit box we manipulate the basis of the USD swap by changing the frequency from 1 year to 6 month and adjusting the number of days per year to get an accurate spread.

Back to top of page

Term Structure Spreads

This graph represents the 30 yr EUR Term structure plotted with the 30 yr USD Term Structure and the spread between these two curves on 9th November 1999.

Back to top of page

Combining Term Structures with Time Series

Horizon II enables you to plot a time series and term structure on the same graph. It is possible to compare the expectation of the market in the past with the reality of it. The graph shows the 3 month forward curve on various dates plotted at various times on the 3 month deposit rate.

Back to top of page