Weighted average but filter?

[Thigs]

I've look at the other weighted average posts and still a little confused so hoping for some help. I have a large data set with SKUs, a metric called pallet factor (how much of that particular SKU fits on a pallet, ex - 54 cases), and a quantity shipped for the month (ex - 32 cases of this SKU shipped in December). 

I need to create a weighted average using the amount of times shipped x the pallet factor, so I'm weighting the pallet factor. 

BUT I need to make it able to be filtered. So if I select the clothes department, it changes the weights to only the clothes department. If it's the pet department, only the pet department, and if its both, then both are included in the weights. 

Here is my formula so far - 

Weighted PF = 

VAR _Table = summarize('Order Data','Order Data'[SKU], "Quantity New", sum('Order Data'[Quantity]), "PF", AVERAGE('Pallet Factor'[PF]))

RETURN

SUMX(_Table, ([Quantity New] * [PF])/sum([Quantity New]))

It is just returning "NaN", so clearly I'm doing something wrong. Are you able to use two different tables in a summarize formula? I've never tried that before. The tables are indirectly related through a fact table. 

[amitchandak · Accepted solution]

@Thigs , try like

Weighted PF =
VAR _Table = summarize('Order Data','Order Data'[SKU], "Quantity New", sum('Order Data'[Quantity]), "PF", AVERAGE('Pallet Factor'[PF]))
RETURN
divide( SUMX(_Table, [Quantity New] * [PF]),SUMX(_Table,[Quantity New]))

or

Weighted PF =
VAR _Table = summarize('Order Data','Order Data'[SKU], "Quantity New", sum('Order Data'[Quantity]), "PF", AVERAGE('Pallet Factor'[PF]))
RETURN
SUMX(_Table, divide([Quantity New] * [PF],[Quantity New]))

[Thigs]

Worked beautifully, thanks a bunch!