You can maybe consider two possible settings: truth-functional logic and predicate logic.
In TFL , an argument consists of two collections of sentences ; the first being the premises, and the second one being the conclusion, and where the premises purport to prove the conclusion, so that a truth assignment in {T,F} can be made of the sentence.
Then maybe your statement is an argument with an empty set of premises, and you can declare it to be either true or false at will.
In Predicate logic, I guess this could be a sentence within an interpretation,where there are predicates for 'hot', for shining, and maybe some non-logical inference rule that tells us what
happens when the sunis shining, and , then, within the interpretation, one can conclude that the sentence is true or false.
Maybe we could also use temporal logic to show the dependence of 'hotness' when the sun shining somehow.