Enhancing the forecasting power of exchange rate models by introducing nonlinearity: Does it work?

Burns, Kelly; Moosa, Imad A.

doi:10.1016/j.econmod.2015.06.003

Cited by 18 publications

(4 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further, the forecasting of exchange rate has been largely conducted on the basis of the structural "models of the 1970s;" Frenkel-Bilson (Frenkel, 1976;Bilson, 1978), Dornbusch-Frankel (Dornbusch, 1976;Frankel, 1979) and Hooper-Morton (Hooper 8 and Morton, 1982) asset models (see Meese and Rogoff, 1983 for more expositions of the models). These theoretical models capture the predictive power of such variables as relative money supplies, relative incomes, short-term interest rate differentials and relative prices among others for exchange rate forecasting (see Moosa and Burns, 2014a;Moosa and Burns, 2014b;Moosa and Burns, 2014c;Burns and Moosa, 2015;Ahmed, et al 2016 for empirical literatures).…”

Section: A Review Of the Literature On Exchange Rate Forecastingmentioning

confidence: 99%

Modelling return and volatility spillovers in global foreign exchange markets

Salisu

Oyewole

Fasanya

2018

Journal of Information and Optimization Sciences

View full text Add to dashboard Cite

The extant literature on exchange rate forecasting on the basis of the Dornbusch-Frankel, Frenkel-Bilson and Hooper-Morton models prominently reveals the dominance of the autoregressive models over the theory-based models. Some studies have however attempted to upturn the results by including the lagged dependent variable in the theory-based models which somewhat implies comparing a modified random walk with a traditional random walk. We follow a different approach both in terms of theory and methodology. We offer an innovative exposition of the Portfolio Balance theory to stock price-exchange rate nexus. Consequently, a predictive model for exchange rate where stock price is a predictor is formulated. The formulated model is expressed in both linear and nonlinear form in order to account for the role of asymmetric changes in stock prices in exchange rate forecasting. Thereafter, we employ the Lewellen (2004) and Westerlund and Narayan (2014) methods which account for any inherent statistical properties of the predictors. Our results validate the Portfolio Balance theory where we show that the sector-level stock prices consistently turn up as good predictors of the exchange rates. The predictive model proposed in this work does not require the inclusion of a lagged dependent variable to beat the autoregressive models which is the practice in the existing literature. We further demonstrate that asymmetry matters to a large extent in the nexus both for the in-sample and out-of-sample predictability.

show abstract

Section: A Review Of the Literature On Exchange Rate Forecastingmentioning

confidence: 99%

Modelling return and volatility spillovers in global foreign exchange markets

Salisu

Oyewole

Fasanya

2018

Journal of Information and Optimization Sciences

View full text Add to dashboard Cite

show abstract

“…For the accurate prediction of exchange rate returns, exchange market literature has been examining the usefulness of various fundamental human and economic variables, while different methodologies and econometric techniques have also been explored (Burns & Moosa, 2015; Cheung et al, 2005; Kilian, 1999; Kilian & Taylor, 2003; Meese & Rogoff, 1983; Pilbeam & Langeland, 2015; Rossi, 2013). Rossi (2013) suggested that traditional fundamentals, such as inflation, interest rates, net foreign assets and output, and money differentials, show fluctuating degrees of out‐of‐sample predictive ability for exchange rate movements (Galimberti & Moura, 2013; Molodtsova & Papell, 2009).…”

Section: Review Of the Literaturementioning

confidence: 99%

Pandemic‐induced fear and government policy response as a measure of uncertainty in the foreign exchange market: Evidence from (a)symmetric wild bootstrap likelihood ratio test

Kathuria

Kumar

2022

Pacific Economic Review

View full text Add to dashboard Cite

The paper aims to examine the ability of a global fear index (GFI) based on the COVID-19 pandemic and government policy responses as a measure of uncertainty in predicting eight Indian rupee-based exchange rate return series: the Australian dollar, the Canadian dollar, the Swiss franc, the US dollar, the euro, the British pound sterling, the New Zealand dollar, and the Japanese yen. The predictability of the daily Indian rupee-based exchange rate return series is tested using the recently developed wild bootstrap likelihood ratio test of Kim and Shamsuddin for the period 2 October 2020 to 8 March 2021. Both symmetric and asymmetric tests revealed GFI as an insignificant determinant of the Indian rupee-based exchange rate return series. However, government policy responses are a significant determinant of the rupee-dollar exchange rate return series.

show abstract

“…However, the study falls into the same trap as some others (e.g. Meese, Rogoff (1983) [1] , Burns, Moosa (2015) [35] , Chinn (1991 [26] , 2010 [25] )) in attempting to explain and suggest that technicalities ("error correction mechanism", "adjustment to disequilibria", "adjustment speed", "non-linearity") somehow have something to do with forecasting exchange rates outside of the known data.…”

Section: No Improvements In Model Outcomes When Compared To Random Walkmentioning

confidence: 99%

“…The study [35] by Burns and Moosa (2015) examines whether introduction non-linearity would improve the "forecasting power" of exchange rate models. We agree with the findings that the "Meese-Rogoff puzzle" cannot be explained by introducing non-linearity, or by introducing any other gimmick into a model.…”

Section: No Improvements In Model Outcomes When Compared To Random Walkmentioning

confidence: 99%

A Simple Model for Expected Exchange Rate E(s) of Currency – It is About the Present Not the Future!

2017

International Journal of Managerial Studies and Research

View full text Add to dashboard Cite

Enhancing the forecasting power of exchange rate models by introducing nonlinearity: Does it work?

Cited by 18 publications

References 48 publications

Modelling return and volatility spillovers in global foreign exchange markets

Modelling return and volatility spillovers in global foreign exchange markets

Pandemic‐induced fear and government policy response as a measure of uncertainty in the foreign exchange market: Evidence from (a)symmetric wild bootstrap likelihood ratio test

A Simple Model for Expected Exchange Rate E(s) of Currency – It is About the Present Not the Future!

Contact Info

Product

Resources

About