The Analysis of Residential Property Price Bubble and Sharia Bank Financing Using MIDAS Regression Model

In economic modeling there are constraints in terms of time frequency differences in the data used as input for estimating a model. One real example is when modeling the relationship between the Islamic banking financing (FIN) and the Housing Price Index (HPI), where the frequency of FIN data is monthly and HPI is quarterly. In this study, it has been tried to develop a model which can be used in the forecasting of Residential Property Price Bubble by using MIDAS (mixed data sampling regression) method which allows the series to be used in the same regression equation from different frequency.Considering the AIC and SIC criteria, it was found that the best performing model out of four alternatives was the weighted equation according to the U-Midas method. The research result shows that the MIDAS specification for modeling the relationship between Islamic bank credit and the property price index is MIDAS with the U-Midas function at lag 4. This model is able to explain variations in property price bubbles of 98.8916%, meaning that the model can be used well for forecasting. Performance the resulting forecast is very good, it is shown that the MAPE value is 0. 2141%.