Forecasting foreign exchange rates with an intrinsically nonlinear dynamic speed of adjustment model

Abstract

Forecasting foreign exchange rates is an important but difficult process; therefore, it is important to use a superior forecasting model. The paper takes up this criterion and proposes to describe and forecast foreign exchange rates by developing an intrinsically nonlinear model with variable and dynamic speeds of adjustment. It is found that the speed of adjusting the random (or expected) to the equilibrium rate is very slow, implying that fiscal policy (statistically insignificat) and monetary policy (statistically significant) may be ineffective to induce changes in the adjustment speed. We also find that the nonlinear dynamic model improves forecasting performance, implying that nonlinearities in the sense of functional forms are exploitable for improved point forecasting of foreign exchange rates.

Currency forecasting based on an error components-seemingly unrelated nonlinear regression model

“…Detailed explanations of equations (1)-(5) can be found in Lin and Chen (1998). Briefly, equation (1) means that the actual change of the spot rate at time t + m, S j (t + m) -S j (t + m -1), is adjusted dynamically, at a dynamic and variable speed, d j (t + m), to the deviation of the desired spot rate at time t + m from the actual spot rate at time t + m -1, S j *(t + m) -S j (t + m -1), plus the random error at time t + m, W j (t + m).…”

“…The error components-seemingly unrelated nonlinear regression (EC-SUNR) model, in which the speed of adjustment is assumed to change variably and dynamically through time, consists of seven equations of which the first five are given by Lin and Chen (1998): …”

“…The most recent development is represented by the study of Lin and Chen (1998). This study is closely related to the development of functional forms.…”

“…The application of VAR models (e.g., Liu et al, 1994) represents another development. One significant development is represented by the studies (e.g., Chinn, 1991;Diebold and Nerlove, 1989;Diebold and Nason, 1990;Lin and Chen, 1998;Meese and Rose, 1991;Brock et al, 1991, among others) that have shown functional forms to be exploitable for improved point forecasting of foreign exchange rates.…”

“…Lin and Chen, 1998). One of them is a technical analysis called the random walk hypothesis (Abuaf and Jorion, 1990;Hakkio and Rush, 1986;Meese and Rogoff, 1983a,b), which is based exclusively on the historical movements of foreign exchange rates without considering the influence of any economic and political factors.…”