Interpolation Methods: A Study of Solving Annual Data into Quarterly and Monthly Data

This study aims to analyse the effect of Export Value (EKS_US$) and Exchange Rate (KURS_US$) on Gross Domestic Product (GDP) as a proxy for economic growth in Indonesia using multiple linear regression based on interpolated monthly data from annual data. In this paper, it is also explained about calculation examples, quarterly and monthly data interpolation methods using data from the Central Bureau of Statistics, namely Indonesia's Gross Domestic Product (GDP) for 2021-2022 at constant 2010 prices. Based on the calculation results, using multiple linear regression of monthly data for the 2018-2022 period, it shows that the estimated model has the best estimator, which is indicated by the variables used, passing the diagnostic test in the form of Multicollinearity, Heteroscedasticity, Autocorrelation, Normality, and Linearity tests. This indicates that the estimator value of the observed data is able to represent the random behaviour of the actual data. However, everything returns to the researcher, at which stage will be chosen is greatly influenced by various factors and considerations of the researcher. Thus, at certain stages and situations, it is very open to researchers to use data interpolation, especially on all data that has flow characteristics, such as GDP data, export values, exchange rates and others. Keywords: Interpolated data; annual data; quarterly data; monthly data; diagnostic tests