Forecasting Foreign Exchange Rate using the Kalman Filter Approach
No Thumbnail Available
Date
2011
Journal Title
Journal ISSN
Volume Title
Publisher
Uva Wellassa University of Srilanka
Abstract
Note: See the PDF Version
The key element to fuel the 'price trend analysis' is the financial tick data (trade data). Tick data has several challenging features of which we intend to consider the most prominent setback when analyzing the price trend, which is data corruption and outliers. This will be our primary concern when building the model. There are different noises found in financial tick data namely process noise, measurement noise and arrival noise. These noises have been broadly studied in the engineering field for the case of an identified deterministic system. The Kalman filter was invented to approximate the state vector of a linear deterministic system in the presence of the process, measurement, and arrival noise. The Kalman filter has been applied in the field of econometrics for the case when a deterministic system is unknown and must be estimated from the data (Lumengo, 2008; Martinelli, 1995).
Many different methods have been offered to deal with signal mining problems in common and trend estimation has also received a great deal of attention especially when the interest is focused on forecasting turning points. In spite of all the differences among methods, one common feature remains in most of them. This is that trends tend to extrapolate themselves into the future as a line with a slope that depends on the recent past information. Although this is an optimal (eg. in a Mean Square Error sense) and a sensible way to progress, it can be systematically erroneous when turning points are at hand. Hence modeling the trend accurately was second principal concern.
The Kalman filter was incorporated in order to reduce the noise in the measurements and to obtain forecasted values of the exchange rates.
Description
Keywords
Economics, Financial Management, Marketing