This is not even logical. Values can Risk for some of the rows and other than Risk for others, then how can you give different column aliases at the same time.
Your problem is that you do not understand what a table is. "What you want is "Automobiles, squids and Lady Gaga" instead of a table. The table is made up of roads; rows are made up of columns. The model scaler values. Each row has one and only one structure; a column does not change from an automobile to a squid depending on some outside influence like your R type. I also worry about the phrase are type; have you ever heard of ISO 11179 naming standards? Is quote our quote a known kind of thing in your trade, or is it metadata that should have no place in a data element name? Should risk and issue be types in a type column instead of trying to change the structure of the table on the fly row by row?