To model the FX spot price of USDCAD, I am using the following data points: GDP TradeAccount CPI Employment Housing etc Many of these categories only have monthly or quarterly data. I'd like a more frequent set - perhaps weekly if at all possible. I'm curious what others have done with this data limitation. One idea is filling-forward the dataset (ie if blank use last value). Not a huge fan of this approach though. What else could be done? Thanks
great point, I will add nat resources into the mix, certainly affects the fundamentals Any insights on how to merge monthly and daily data? My options are: Convert monthly to daily, using same value throughout the month. Sum daily data to monthly level and proceed. Looses granularity but might be the easiest to start with