Purpose: The purpose of this paper is to estimate long-term currency exchange rate and also identify the key factors for decision makers in the currency exchange market. The study is expected to aid decision makers to take positions in the dynamic Forex market. Design/methodology/approach: This study is based on quantitative and fundamental analysis of statistically oriented regression models. The trend of quarterly exchange rates is investigated using 110 variables including economic elements, interest rate and other currencies. This research is based on the same information that banks’ dealers use for the analysis. Ordinary least squares linear regression also known as “least squared errors regression” was used to estimate the value of the dependent variable. Findings: The study concludes that “only Australian economic data” or “only the US economic data” cannot fully reflect the trend of AUD/USD. EUR influences AUD relatively larger than the other main market currencies. Six-month Australian interest rate itself affects AUD/USD trend much more than the six-month interest difference between AUD and USD. Research limitations/implications: The results indicate that the economic autoregressive moving average model can be used to predict future exchange rate using primary factors identified and not from the generic market or economic view. This helps adjust to the general, common (and possibly wrong) views when making a buy or sell decision. Originality/value: This is one of the first studies in the context using the information of bank dealers for AUD/USD. This study is highly relevant in the current context, given the significant growth in Forex trade. © 2019, Emerald Publishing Limited.