Score aggregation results in ElasticSearch

Question

I created a type called test , which consists of a nested price field .

Each document may consist of several prices, and each price consists of a price and a date :

...
"prices":[{  
      "date" : "2017-02-08T16:04:20.424Z",
      "price" : 120
   },
   {  
      "date" : "2017-02-08T17:35:01.424Z",
      "price" : 70
   }]
...

I am trying to create a search query that first aggregates the prices of each document to the minimum price , and then dials all the minimum values ​​returned from the aggregation using the linear decay function

Currently, I have an aggregation request for all documents:

{
    "query": {
        "match_all": {}
    },
    "aggs" : {
        "prices" : {
            "nested" : {
                "path" : "prices"
            },
            "aggs" : {
                "min_price" : { "min" : { "field" : "prices.price" } }
            }
        }
    }
}

This returns the minimum price in the test .

, :

{
  "query": {
    "nested": {
        "path": "prices",
        "query": {
            "function_score": {
              "functions": [
                {
                  "linear": {
                    "prices.price": { 
                      "origin": "50", 
                      "offset": "50",
                      "scale":  "20"
                    }
                  }
                }
              ]
            }
        }
    }
  }
}

. , , , .

?